Derivative Markets

in India
Trading, Pricing, and Risk Management
 Derivative Markets
in India
Trading, Pricing, and Risk Management

Alok Dixit
Assistant Professor
Finance and Accounting Area
IIM Lucknow

Surendra S. Yadav
Professor of Finance
Department of Management Studies
IIT Delhi

P.K. Jain
Professor of Finance
Department of Management Studies
IIT Delhi

Tata McGraw Hill Education Private Limited

NEW DELHI
McGraw-Hill Offices
New Delhi New York St Louis San Francisco Auckland Bogotá Caracas
Kuala Lumpur Lisbon London Madrid Mexico City Milan Montreal
San Juan Santiago Singapore Sydney Tokyo Toronto
Published by Tata McGraw Hill Education Private Limited,
7 West Patel Nagar, New Delhi 110 008
Copyright © 2012, by Tata McGraw Hill Education Private Limited
No part of this publication may be reproduced or distributed in any form or by any means,
electronic, mechanical, photocopying, recording, or otherwise or stored in a database
or retrieval system without the prior written permission of the publishers. The program
listings (if any) may be entered, stored and executed in a computer system, but they may
not be reproduced for publication.
This edition can be exported from India only by the publishers,
Tata McGraw Hill Education Private Limited.
ISBN (13): 978-0-07-133295-8
ISBN (10): 0-07-133295-2
Vice President and Managing Director—Asia-Pacific Region: Ajay Shukla
Asst. Sponsoring Editor: Simanta Borah
Sr. Copy Editor: Neha Sharma
Manager—Production: Sohan Gaur
AGM—Sales and Business Development: S Girish
Asst. Product Manager—Business & General Reference: Priyanka Goel
General Manager—Production: Rajender P Ghansela
Manager—Production: Reji Kumar
Information contained in this work has been obtained by Tata McGraw-Hill,
from sources believed to be reliable. However, neither Tata McGraw-Hill nor
its authors guarantee the accuracy or completeness of any information published
herein, and neither Tata McGraw-Hill nor its authors shall be responsible for any
errors, omissions, or damages arising out of use of this information. This work is
published with the understanding that Tata McGraw-Hill and its authors are supplying
information but are not attempting to render engineering or other professional
services. If such services are required, the assistance of an appropriate professional
should be sought.
Typeset at Shubham Composer, WZ-437 Madipur Village, New Delhi 110063 and printed at
Rajkamal Electric Press, Plot No. 2, Phase IV, HSIIDC, Kundli, Sonepat, Haryana - 131028
Cover Printer: Rajkamal Electric Press
Cover Designer: Kapil Gupta
RLYCRRZHDZBAQ
 To
The Almighty
and
Our Family Members
 Preface

Financial Derivatives, amongst their other functions, provide a variety of tools

for managing/tailoring risk for all types of investors in a financial market.
It is empirically observed that many equity funds are created based on the
practice called ‘indexing’. Indexing is an investing style wherein an equity
portfolio is benchmarked to the leading index of the economy. And, therefore,
the portfolio is expected to be in strong association with the benchmarked
index. The observed high correlation in the portfolio returns with those of
the benchmarked index makes index derivatives a natural choice for hedging
risk of equity portfolios.
Amongst the index derivatives, index options play an important role in
the economy as they provide a better hedging mechanism to the investors
compared to index futures. The perceived superiority of index options is
generally observed in the times of global uncertainty when the markets
experience periods of high volatility. The options contracts, unlike futures,
allow fund managers to take advantage of favourable movements in the
market along with the protection against the unfavourable movements. In sum,
these financial instruments help investors to ensure stable earnings tailored
to their risk appetite and, therefore, facilitate in mobilization of funds from
the domestic as well as foreign investors.
Empirically, the performance of options market at its well identified
functions, namely, risk hedging, price discovery and enhancement of
liquidity in the underlying’s market depends, to a greater extent, on its
efficiency (Ackert and Tian, 2000). Consequently, the efficiency of options
market has been of equal interest to the academics as well as practitioners
and a number of studies have been carried out across the globe in different
options markets.
In view of the perceived advantages of such financial innovations and
the nascent stage of derivatives market in India, it is desired to carry out a
comprehensive study to examine the pricing efficiency of the index options
market. In response to this and lack of adequate literature on the subject in
the context of Indian options market, the present study attempts to assess the
viii ● Preface

efficiency of the S&P CNX Nifty index options, the options which are traded
on the leading index of the National Stock Exchange (NSE).
The issue of market efficiency has been dealt with by using a comprehensive
research methodology. The methodology adopted consisted of a two-pronged
approach which built on the analysis of secondary as well as primary data
to ascertain the state of index options market in Indian. The secondary data
analysis has been carried out using select ‘model-free’ as well as ‘model-
based’ approaches. For the purpose, the study used the data for a period of six
years from June 04, 2001 (starting date for index options in India) to June 30,
2007. Moreover, a survey amongst the trading member organizations based
at National Capital Region (NCR) and Mumbai has also been conducted to
corroborate the findings from the secondary data analysis.
The ‘model-free’ approaches to assessing the options market efficiency
include the test of two popularly known conditions on options prices, namely,
(i) test of lower boundary conditions on the options prices and (ii) test of
put-call parity relationship. Likewise, the ‘model-based’ approaches have
dwelt upon, (i) test of rational expectations hypothesis on the term structure
of implied volatilities and (ii) examination of the informational efficiency of
implied volatilities vis-à-vis estimates from the select conditional volatility
models. In addition, the survey assessed the opinion of respondents on five
major dimensions relating to the state of options prices in Indian derivatives
market. These dimensions include: (i) level of usage of the options; (ii)
understanding of put-call parity relationship; (iii) awareness and use of
models for options valuation; (iv) correctness of options pricing, its impact,
and existence and exploitability of arbitrage opportunities; and (v) need of
regulations and educational initiatives for the betterment of the market.
Empirically, the findings of the ‘model-free’ approaches have been
corroborated with those of ‘model-based’ approaches. One of the notable
findings is that put options market is more inefficient compared to call options
market. Precisely, the put options have been overpriced in relation to call
options. The lopsided pattern (more inefficiency in the case of put options vis-
à-vis call options) of violations could be attributed to ‘short-selling constraint’
in Indian securities market during the period under reference. This is borne
out by the fact that the arbitrage strategy requires taking a short position in
the underlying asset to correct such anomalies, i.e., overpricing of put options.
And, it may be noted that the short-selling was not allowed in Indian market
during the period under reference.
However, it is generally argued that futures market can ideally be used in
case short-selling is banned in the market given that the futures are available
on the underlying asset. In view of this, the model free approaches were also
 Preface ● ix

tested using S&P CNX Nifty futures prices. The major findings of the analysis
show that the futures market could identify only some of the mispricing
signals as the futures themselves were underpriced. The underpricing of
futures can be traced to the absence of short-selling facility in the market
as the correction of under-pricing of futures requires short-selling in place.
Therefore, it would be reasonable to conclude that the short-selling constraint
has emerged as one of the major reasons for the typical pattern of violations
in Indian options market.
Another notable finding of the research is that a majority of violations both
in the cases of call and put options could not be exploited on account of the
dearth of liquidity in such contracts. An adequate level of liquidity in options
as well as underling’s market is needed to execute the appropriate arbitrage
strategy successfully. The lack of liquidity causes a risk called immediacy-risk;
that is, inability of the arbitrageur to execute the strategy within the required
timeframe. Failing which, the observed arbitrage profits could turn into losses.
Besides, higher bid-ask spreads are observed for the contracts having low
liquidity and vice versa. And, therefore, the lower level of liquidity reduces
the observed profits as well.
In nutshell, the major findings presented in this book reveal that the Indian
investors are not exhibiting rational behaviour while valuing index options
during the period under reference. This is indicative of price inefficiency in
index options market in India. The revealed state of options pricing in the
derivatives market can be attributed to the short selling constraints, dearth
of liquidity and lack of proper understanding of the market amongst market
participants/investors.
In addition, a brief sketch of the recent initiatives taken by SEBI and their
possible impact on the market efficiency has been delineated in the book.
Besides, the need of further initiatives warranted to restore efficiency of the
options market in India has been given in the form of recommendations.
This book is expected to be useful for the academic community (especially
MBA/M.com students and the researchers who are working in the area of
Futures and Options), industry, the regulatory body (SEBI) and the major
exchanges of the country. The book can also be used for the class room teaching,
especially to explain the arbitrage mechanism in options market. In addition,
the book provides insight into the future research that can be attempted in
the domain of Futures and Options market. Besides, the regulators and stock
exchanges of the other countries which have recently introduced the Futures
and Options in their markets may find this book useful.
 Acknowledgements

At the outset, we would like to thank the Almighty for His blessings to inspire
us to accomplish this academic endeavour. This work has been possible
because of the help, encouragement, cooperation and guidance of many people
and we convey our heartfelt thanks to all of them.
We are grateful to Prof. Surendra Prasad, Ex-Director, IIT Delhi, Prof.
R.K. Shevgaonkar, Director, IIT Delhi, and Prof. Devi Singh, Director, IIM
Lucknow for their kind cooperation and support. We are also thankful to all
the colleagues in Department of Management Studies, IIT Delhi, and in IIM
Lucknow for their good wishes for this endeavour.
In addition, we are deeply obliged to all the respondents of the survey, who
took out time from their busy schedules to provide data for this work.
Dr. Alok Dixit takes this opportunity to express his deepest gratitude to
respected guides (gurus) and co-authors, Prof. Surendra S. Yadav and Prof.
P.K. Jain, for their valuable guidance, inspiration, motivation, and untiring
efforts in completion of this project.
Last but not the least, we are thankful to our family members and loved
ones for their continuous encouragement and support.
A D
S S. Y
P.K. J
 Contents

Preface vii
Acknowledgements xi
Abbreviations xv
Units of Measurement xvii

C  1 Financial Derivatives 1
1.1 Introduction 1
1.2 Types of Financial Derivatives 2
1.3 Use of Financial Derivatives 7
1.4 Factors Contributing to the Growth of Derivatives 7
1.5 Evolution of Derivatives Market in India 8

C  2 Equity Options and Risk Management 12

2.1 Introduction 12
2.2 Trading Strategies Using Option for Profit Making
(Speculation) 14
2.3 Use of Options for Hedging Risk 27

C  3 Testing Lower Boundary Conditions for the S&P

CNX Nifty Index Options 31
3.1 Introduction 31
3.2 The Lower Boundary Conditions 33
3.4 The Data 42
3.5 Analysis and Empirical Results 44
3.6 Concluding Observations 61
End Notes 62

C  4 A Test of Put–Call Parity Relationship on S&P CNX

Nifty Index Options Market 63
4.1 Introduction 63
4.2 Put–Call Parity (PCP) Relationship 65
4.3 Data 75
xiv ● Contents

4.4 Analysis and Empirical Results 76

4.5 Concluding Observations 95

C  5 Testing the Expectations Hypothesis on the Term

Structure of Volatilities Implied by S&P CNX Nifty
Index Options 97
5.1 Introduction 97
5.2 Expectations Hypothesis on the Term Structure of
Implied Volatilities 98
5.3 Data and Estimation of Implied Volatility 102
5.4 Analysis and Empirical Results 107
5.5 Concluding Observations 120
End Notes 120

C  6 Informational Efficiency of Implied Volatilities of

S&P CNX Nifty Index Options 123
6.1 Introduction 123
6.2 Characteristics of Volatility and Conditional Volatility
Models 124
6.3 Data and Implied Volatility 129
6.4 Informational Efficiency of Implied Volatility
Vis-à-vis Volatilities Estimated Using ARCH
Volatility Models 131
6.5 Analysis and Empirical Results 133
6.6 Concluding Observations 149
End Notes 150

C  7 Survey Analysis and Findings 151

7.1 Introduction 151
7.2 Survey Methodology 151
7.3 Profile of Respondents in Terms of Their Background
and Trading Volume  155
7.4 Analysis and Empirical Results 156
7.5 Concluding Observations 176

Glossary 179

References 183
 Abbreviations

AACF Autocorrelation Function

AIC Akaike Information Criterion
AOM Australian Options Market
ARCH Autoregressive Conditional Heteroscedasticity
ARIMA Autoregressive Integrated Moving Average
BSE Bombay Stock Exchange
CBOE Chicago Board Options Exchange
CME Chicago Mercantile Exchange
EGARCH Exponential Generalized Autoregressive Conditional
Heteroscedasticity
EOE European Options Exchange
F&O Futures and Options
FOM Finnish Options Market
GARCH Generalized Autoregressive Conditional Hetero-
scedasticity
GED Generalized Error Distribution
IV Implied Volatility
LBC Lower Boundary Condition
LIFFE London International Financial Futures Exchange
LR Likelihood Ratio
LTOM London Traded Options Market
NSE National Stock Exchange
OSE Oslo Stock Exchange
OTC Over The Counter
PACF Partial Autocorrelation Coefficient
PCP Put-Call Parity
PHLX Philadelphia Stock Exchange
SIC Schwarz Information Criterion
VIX Volatility Index
VXN CBOE Nasdaq Volatility Index
VXO CBOE OEX Volatility Exchange
 Units of Measurement

1 Lac = 100 thousand

10 Lac = 1 million
1 Crore = 10 million
100 Crore = 1 billion
1000 Crore = 10 billion
1 C H A P T E R

Financial Derivatives

1.1 INTRODUCTION
Financial markets play a vital role in capital creation in an economy. Precisely,
these markets are essential for raising capital through long-term sources of
finance, for example, equity shares and debentures/bonds. This is in view of
the fact that these markets not only facilitate in raising the fresh capital but
also ensure the trading of the investment vehicles. Trading in the investment
vehicles provides flexibility to the investors to liquidate their investments
virtually any time (to be on the safer side, any day) they want. However,
financial markets are largely subject to market risk. Market risk represents
summative effect of the economic forces (inflation, exchange rate, GDP
growth, etc.), which are unpredictable to a marked extent, and could cause
substantial alteration in the value of the assets traded in the open market. In
view of this, an investor (hedger) in the market looks for some instrument
that could provide him protections against such risk. At the same time, there
are participants (speculators) in the market who are willing to take this risk.
In a financial market, derivative instruments work as link between the two
entities; namely hedger (one who wants to transfer its risk) and speculator (the
other willing to take this risk). Besides, derivatives instruments are identified
with their contribution in terms of price discovery and increased liquidity
in the underlying’s market. In sum, derivative instruments are expected to
provide three major benefits in a financial market, viz. (i) transfer of risk
from one party to another party (hedging), (ii) better price discovery in the
underlying’s market and (iii) increase in liquidity of the underlying asset. In
turn, these benefits lead to better capital allocation in an economy.
Derivatives, also known as financial innovations, are defined as those
instruments that derive their value from the changes in the value of the
underlying asset; that is, in itself it does not have any value, unlike stock and
bonds. Rather, their value depends on typical characteristics/ behaviour of the
underlying asset. The underlying asset can be a financial asset (e.g. individual
2 ● Derivative Markets in India

shares, bonds and indices thereof), commodity (e.g. grains, metal, etc.) or
reference rate (e.g. interest rate, exchange rate, etc.). In their simplest form,
known as plain-vanilla instruments, derivatives include Forward Contracts,
Futures, Options and Swaps. Moreover, over the period of time, with the
increased complexity of financial markets, a variety of advanced derivative
instruments have been developed. Some of such instruments are Swaptions,
Barrier Options, Collaterlised Debt Obligations (CDOs), Collaterlised Bond
Obligations (CBOs), Credit Default Swaps (CDS), etc.
Financial derivatives, in particular, are those instruments that have financial
assets as the underlying. Naturally, these instruments become an obvious
choice for managing/tailoring financial risk of the assets being traded in the
market. Amongst financial derivatives, the instruments that are based on
indices (e.g. index futures and index options) serve as a natural choice to the
portfolio (fund) managers for tailoring the market exposure of funds based
on the risk appetite of their clients.

1.2 TYPES OF FINANCIAL DERIVATIVES

With increased complexity of financial markets and demand from different
types of investors, a variety of financial derivatives have been developed. Some
of the most popularly known basic financial derivatives include Forwards,
Futures, Options and Swap contracts. Based on one of the most common
classifications offered to derivatives contacts, these are classified in two
major categories: Exchange-traded and Over-the-counter (OTC) derivatives.
Exchange-traded derivatives, as evident from the name, are those standardized
products that are traded on an exchange; e.g. Futures and Options contracts.
In contrast, OTC derivatives are those derivative contracts that are trailed
products (usually a bilateral agreement to trade the underlying at a future
trade) and are not traded on any exchange. Some of the most common examples
of OTC derivatives products are Forward and Swap contracts.
Another popular classification of derivatives is drawn based on the
underlying asset. For example, Equity derivatives – derivatives on shares and
indices thereof (NSE Nifty in the context of Indian capital market), namely
stock futures, stock options, index futures and index options; Foreign Exchange
(Forex) derivatives – those based on foreign exchange rate or involves exchange
of sums denominated in more than one currency, e.g. currency forwards,
currency futures (US Dollar and Indian National Rupee futures, USDINR
futures contract available in Indian capital market), currency options and
cross-currency swaps, etc.; Credit derivatives – instruments that are designed
to provide shield against the credit/default risk, viz. CDS, CBO, CDO, etc.;
Interest rate derivatives – these instruments involve promise to invest/borrow
certain amount in future at the interest rate locked-in now, e.g. interest rate
futures, interest rate swaps, etc.
 Financial Derivatives ● 3

Having discussed some major categories of financial derivatives, let us

move on to the discussion of some of the most commonly known financial
derivatives, viz. Forwards, Futures, Options and Swap contracts.

1.2.1 Forward Contracts

A forward contract is an agreement between two parties to transact (buy/
sell) the underlying asset in future at a certain price, which is decided now.
A forward contract is an OTC derivatives instrument, and therefore, it is a
tailored instrument negotiated by the two parties involved in the contract. Such
contracts are not traded on any exchange, and therefore, it is not easy (nearly
impossible) for either party to square-off its position before the maturity of
contract. Forward contracts are more commonly traded in the foreign exchange
market and fixed income securities market. In forex market, these are used to
hedge the foreign exchange risk, in general, and transactional risk, in particular.
Likewise, in fixed income securities market, these instruments help in hedging
against the interest rate risk (e.g. Forward rate agreements, FRA).

Example 1.1 Suppose an Indian firm exported some goods worth 1,00,000

Euros (€) to a French firm. The French firm promised Indian firm to pay three
months from now. Further, suppose that the Euro is currently traded at ` 58.50/
58.80 (per €) and the Indian firm expects that Euro will weaken in due course.
The intuition of depreciation of foreign currency gets corroborated when it looks
at the forward rates quoted by the forex dealers. These quotes are as under:

Spot rate 1 month forward 2 months forward 2 months forward

(`/ €) Rate Rate Rate
(`/ €) (`/ €) (`/ €)
58.50/80 57.10/40 56.50/80 55.50/80

The exporter can make use of forward market by selling Euros 3-months
forward in the market. This can be achieved by entering into a 3-months forward
contract, and therefore, the exporter can avoid the foreign currency exposure
due to unfavourable movements in Euro by selling it at ` 55.50 per Euro. Now,
whatever happens to forex rates three months from now, he will receive a certain
amount, i.e., ` 55,50,000 (1,00,000 ¥ ` 55.50).

1.2.2 Futures Contracts

A futures contract, alike forward contract, is an agreement between two parties
to buy/sell the underlying asset at a future date at a price which is decided
now. Amongst other differences, the major characteristic that differentiates
futures from forwards is that these are exchange-traded instruments. Given
the fact that futures are traded on an exchange, these are no longer tailored
products; rather, these are standardized contracts in terms of contract size
and maturity date. Another important feature of futures contract is that, as
4 ● Derivative Markets in India

a risk containment measure, these instruments require some initial margin

to be deposited with the exchange to take a position in the market. Further,
the margin amount keeps on fluctuating due to ‘marking to market’ feature
of futures contracts. Marking-to-market connotes daily settlement of futures,
i.e., loss (gain) generated over the previous day closing price of the futures is
debited (credited) in the respective accounts of parties. Once the margin money
goes below a specified limit due to such fluctuations, a margin call is made.
Major application of futures contracts are seen in Equity market, viz. Futures
contracts on individual shares and those on indices of such stocks. Besides,
these instruments are also traded in fixed income securities market, as they
provide shield against the interest rate risk (Interest Rate Futures). Futures
serve as a tool for hedging exposure in the financial markets. Besides, these
are heavily used for speculating in the market as well.

Example 1.2 For example, an investor holds 1000 shares of TATASTEEL,

which are currently traded at ` 610. He fears that the market may go down
in next one month’s time, and therefore, he wants to hedge himself against
the likely adverse movements in the market. Suppose that a futures contract
on TATASTEEL scheduled to expire one-month from now is traded at ` 610.
One futures contract includes 100 shares of TATASTEEL. Therefore, the
investor needs to sell 10 futures contracts short in the market to hedge against
market risk. For taking this position, he will be required to deposit the margin
amount as notified by the exchange. In the meantime, his futures position will
be ‘marked to market’ on daily basis (i.e. his account will be debited {credited}
if the futures price rises {declines} compared to the previous day’s future price)
up to maturity of the contract.
At maturity date, suppose the TATASTEEL’s share as well as futures
contract thereof close at ` 5801. The investor will end up making a loss of `
30,000 on the spot market; at the same time, he will make a gain of ` 30,000 on
the futures leg of the transaction. In sum, loss on spot market will be fully offset
by the gain on futures leg of the transaction, in case he closes out his position
in the market. Had he not taken any position in futures market, he would have
made a loss of ` 30,000.

1.2.3 Options Contracts

Options are the contracts that give a right to its holder to buy/sell the underlying
asset at a specified price (called as strike/exercise price) on or up to the maturity
date of contract. It is important to note that options, unlike futures, do not entail
any obligation to buy/sell the underlying asset when market turn out to be
1
If the market is efficient, we expect that the closing price of the futures contract should be the
same as that of the underlying’s price. That is, both the prices should converge at maturity date
of the contract.
 Financial Derivatives ● 5

favourable. That is, these instruments protect an investor against unfavourable

movements in the market; at the same time, offer the opportunity to tap the
favourable movements in the market as well.
Depending on timing of exercising the contract, these instruments are put
to a dichotomous classification, namely European options and American
options. European options are those contracts that give a right that can be
exercised only at maturity, whereas American options are the contracts that
can be exercised anytime up to maturity. Another important classification of
these contracts is made based on the right that they provide, i.e., to buy/sell
the underlying asset. An options contract which gives a right to its holder to
buy the asset is called a call options, and the one which offers a right to sell the
underlying asset is called a put option.
In the case of options, it is important to note that these are traded as OTC
as well as exchange-traded instruments. In an option contract, be it OTC or
exchange-traded, there are two parties involved, namely, holder of the option
and writer of the option. Holder of option is the investor who buys the right
to buy/ sell the underlying asset. In contrast, the writer of the options is an
investor who promises to sell/buy the underlying asset at the strike price
(irrespective of the market condition), in case the contract is exercised by the
holder of the option. He is obliged to honour his position once the contract
is assigned. Naturally, holder of option has to compensate writer for the risk
that he takes by promising to buy/sell at fixed price irrespective of the market
situation at maturity. The compensation that an option writer gets is called
option premium. The option premium (price of the options contract) depends
on certain factors as pointed out by Fisher Black and Myron Scholes in 1973.
They developed a model, popularly known as Black–Scholes (1973) model—a
milestone in finance, by using these factors to value an option contract. These
factors include, strike price, spot price, volatility of the underlying, maturity
time and risk-free rate of return. Further, Merton (1973) added another factor
to this list, i.e., dividend (yield/ absolute amount), if any, declared during the
maturity period of the options contract.
Options are very useful instruments for portfolio managers, especially
when the financial markets are experiencing very high volatility. However,
their speculative appeal to trades is relatively less compared to that in the
case of futures contracts. Most common application of options contract is
seen in Equity market (stock and index options) and Forex markets (currency
options).

Example 1.3 In addition to data given in Example 1.2, further suppose that an
option contract on TATASTEEL, scheduled to expire one-month from now and
with the strike price of ` 610, is currently traded at ` 30 per contract. Say, the
contract size is 100 shares. Now suppose that the investor is willing to choose options
route to hedge against the unfavourable movements in the market. He needs to buy
6 ● Derivative Markets in India

10 put option contracts (Total no. of shares/No. of shares per contract, 1000/100),
and it will cost him ` 300 upfront.
At maturity (in the case of European options) or up to maturity (in the case
of American options), if the price of TATASTEEL shares goes down below `
610, the investor can exercise the option and can sell these shares at ` 610. For
example, the options that the investor purchased are European in nature and
at maturity TATASTEEL closes at ` 580. The investor will incur a loss of
` 30000 on his position in the spot market. At the same time, he will exercise the
option and sell the shares at ` 610. Therefore, he will be able to recover ` 30000
from his position in options market and his effective loss will be the premium
that he paid upfront, i.e., ` 300.
It is important to note that he would have done the same by using futures as
well (refer to Example 1.2). However, the advantage of option is that investor has
the flexibility not to sell his shares at ` 610 when market values TATASTEEL at
more than ` 610, say ` 640. In contrast, in the case of futures, he has to sell the
shares at ` 610 even if the market value of share turns out to be more than ` 610.

1.2.4 Swaps Contracts

Swap, an OTC derivative instrument, is an agreement between two parties to
exchange cash flows (interest or principal cum interest) for a specified period
of time. Swaps are most commonly used in forex and fixed income securities
markets. Some common applications of swaps are seen in terms of currency
swaps and interest rate swaps. A currency swap is an agreement between
two parties to exchange cash flows denominated in different currencies for a
specified period of time. An interest rate swap involves exchange of interest
amounts between two parties for a stipulated time period, wherein interest
amounts are calculated based on fixed and floating interest rates for the two
legs of swap transaction, respectively. The amount of interest exchanged
between the two parties is calculated on a notional principal amount. In the case
of interest rate swap, it is important to note that, in general, principal amount
is not exchanged between the parties. Normally, such transactions are routed
through an intermediary financial institution.

Example 1.4 Suppose a firm wants to raise some capital through debt market.
The firm finds that it can raise the desired amount based on floating interest
rate, i.e., MIBOR2 plus 100 basis points; however, it is interested in getting it
exchanged for a fixed interest rate loan for whatsoever reason (may be the firm
is new and does not want to take this risk). At the same time, there is another
firm that could raise the debt amount at the fixed rate of interest, say at 8% p.a.,
but was willing to exchange it for floating rate of interest. Both such firms can
explore the possibility of swapping their respective interest liability through the
2
MIBOR stands for Mumbai Inter Bank Offered Rate. This is a reference rate used for lending
and borrowing among banks.
 Financial Derivatives ● 7

financial intermediary dealing in swap agreements for a suitable time period

and, therefore, can have access to desired loan scheme, i.e., fixed/floating interest
rate. Once the deal is finalized for a stipulated time period, the interest liabilities
are routed through the financial intermediary and the intermediary charges its
commission from both the parties for the services provided. In this regard, it
is important to note that only interest liabilities are exchanged and principal
amounts remain with the original parties.

1.3 USE OF FINANCIAL DERIVATIVES

The use of financial derivatives can be classified in three major categories, namely
(i) hedging the financial risk or financial risk management, (ii) speculating
through different profit strategies and (iii) as ex-ante forecast of underlying’s
value.
Hedging refers to process of managing risk/variations in the value of the
asset or portfolio of assets. In hedging, an investor tries to confine expected
variations in the value of asset within a tolerable limit (may be set by himself or
imposed by clients) by taking suitable position in the derivatives market. Some
of the examples of hedging have already been discussed in the previous section
while enumerating types of financial derivatives. Another use of derivatives is
made to generate profit by speculating on direction and magnitude of direction
in the market. Speculation essentially involves the views/take of the investor
on the direction of market and magnitude thereof in near future. Based on
his expectation about the future, the speculator tries to generate profit in the
derivatives market by taking a suitable position/positions in the market. It is
important to note that a speculator is always exposed to risk, as his expectation
may well be reversed, i.e., market may turn just opposite to his expectations.
Both hedging and speculation in derivatives market, especially using equity
derivatives market, have been covered in Chapter 2.
The third use of derivatives market is to generate ex-ante forecast of the
underlying’s value. There have been many studies which attempted to test the
same across the globe in different financial markets. The results suggest that
derivatives market perform as a good predictor of future’s value. For example,
in majority of developed financial markets, interest rates implied from interest
rate futures and volatilities implied from equity options have been found to be
superior forecast of future vis-à-vis those generated by historical data based
models. The forecasting ability of Indian index options market (in terms of
forecasting volatility) has been thoroughly examined in Chapter 6.

1.4 FACTORS CONTRIBUTING TO THE GROWTH OF

DERIVATIVES
Financial derivatives play a variety of roles in a financial market, from restoring
stability in the market to better price discovery. By virtue of functions they
8 ● Derivative Markets in India

perform, derivatives have witnessed phenomenal growth in Indian capital

market. The same holds true for global scenario as well. Some of the major
factors that help derivatives grow in a financial market are as follows:
∑ Financial derivatives help in transfering of risk from one party to another
party and, therefore, are expected to increase participation in the financial
market, which in other words, leads to better capital allocation. These
instruments make financial market a more viable option for a variety
of investors who, otherwise, might not have participated in the market.
For example, a risk-averse investor, in presence of derivatives products
in a market, has an option to transfer risk once volatility in the market
hits a level beyond his tolerance. Similarly, it provides an opportunity
to risk-takers as well; this is in view of the fact that participants can
take leveraged positions, i.e., they can achieve desired exposure in the
market with relatively very less initial investment (compared with spot
market). In sum, it is expected to increase participation in the market
and, therefore, expedite the process of capital creation in an economy.
∑ Another important factor that has contributed towards growth of
derivatives is cross-border integration of economies. Since economic
integration of countries, amidst variety of benefits, transfers various risks
of from one economy to another, it results in increased risk (volatility) in
the financial markets of integrated economies. Since derivatives provide
a shield against such risks, growth in this segment of financial markets
(derivatives) becomes imperative.
∑ Derivatives help portfolio managers to provide their clients with a
variety of structured financial products, which are best suited to their
risk appetite. For example, an investor demands for an investment
opportunity with very high risk and return profile. Such a strategy
can be developed using the spot market itself; but the manager would
prefer a route through derivatives market, given the less amount of
investment required in latter to have same level of exposure. Therefore,
the usage of derivatives products in serving variety of investors (through
structured financial products), amongst other benefits, leads to growth
of this segment of financial markets.
In sum, the perceived benefits of derivatives to participants in a financial
market, clubbed with cross-border economic integration, can be designated
as major building blocks for the growth of derivatives in an economy. Per-se,
this seems true for Indian derivatives market as well.

1.5 EVOLUTION OF DERIVATIVES MARKET IN INDIA

In view of the empirically established advantages of derivatives and in
response to the long-felt need of hedging facility in the Indian securities
market, the derivatives market officially took off in the Indian securities market
 Financial Derivatives ● 9

on June 11, 2000, with the launch of futures contracts on the leading index
of National Stock Exchange (NSE) and Bombay Stock Exchange (BSE). The
derivatives were started in the Indian market in response to the report of L.C.
Gupta Committee. The committee conducted a survey amongst the market
participants to gauge the need of derivatives market in India and found that
there has been a well-awaited need of having a derivatives market for hedging
their financial risk. The derivatives segment in Indian securities market has
registered remarkable growth in its nearly 11 years of existence. The data on
the growth of derivatives segment are summarized in Table 1.1 and Fig. 1.1.
It may be deciphered from the data (Table 1.1 and Fig. 1.1) that the
derivatives market has experienced a whopping growth, evident from the
current volumes which amounts to many times of its initial trading volume
in the very first year of its existence.
Another notable observation from the data on the growth of derivatives
market is that the majority of trading in derivatives market has been seen in
the futures segment compared to that in the options market. The reason behind
this could be the ease of trading in futures market as options are comparatively
difficult to understand, and moreover, the market has got less experience of
the derivatives market, as it has been launched recently in Indian securities
market. Some facts on the development of derivatives market in India along
with its growth story has been summarized in following sub-section.
Total turnover of derivatives market in India, 2000–01 to 2008–09

Total turnover of derivatives market

Figure 1.1 Trading volume of Indian derivatives market (in rupees) from 2000–01 to 2008–09

1.5.1 Some Stylised Facts about Indian Derivatives Market

∑ Futures on the indices of BSE and NSE (BSE Sensex and NSE Nifty,
respectively) were allowed as a first derivative in India in June, 2000.
∑ Likewise, Index options on indices of BSE and NSE (BSE Sensex and NSE
Nifty, respectively) were first allowed for trading in June 04, 2001.
∑ Options on individual stocks were introduced in July 2001.
∑ Futures on individual stocks, a peculiar feature of Indian derivatives
market, were started in November 2001.
 10
●

Table 1.1 Business Growth in Indian Derivatives Segment, 2000–01 to 2008–09

Year Index futures Stock futures Total turnover Index options Stock options Total turnover Total turnover of
Derivative Markets in India

of futures of options segment derivatives market

market (F&O segment)
2000–01 2,365 0 2,365 0 0 0 2,365
2001–02 21,483 51,515 72,998 3,765 25,163 28,928 1,01,926
2002–03 43,952 2,86,533 3,30,485 9,246 1,00,131 1,09,377 4,39,862
2003–04 5,54,446 13,05,939 18,60,385 52,816 2,17,207 2,70,023 21,30,610
2004–05 7,72,147 14,84,056 22,56,203 1,21,943 1,68,836 2,90,779 25,46,982
2005–06 15,13,755 27,91,697 43,05,452 3,38,469 1,80,253 5,18,722 48,24,174
2006–07 25,39,574 38,30,967 63,70,541 7,91,906 1,93,795 9,85,701 73,56,242
2007–08 38,20,667 75,48,563 1,13,69,231 13,62,111 3,59,137 17,21,247 1,30,90,478
2008–09 35,70,111 34,79,642 70,49,753 37,31,501 2,29,226 39,60,728 1,10,10,482
Source: www.nse-india.com
 Financial Derivatives ● 11

∑ Interest rate futures took off in June 2003.

∑ In India, only equity and currency Options are being traded as of now.
∑ Presently, Options and Futures are available on more than 230 individual
stocks and 8 indices.
∑ Trading volume of derivatives market has registered a whopping
increase from ` 2.37 thousand crores in 2000–01 to ` 110.10 lakh crores
in 2008–09.
∑ Turnover of Index Options in Indian market rose from ` 3.8 thousand
crores in 2001–02 to ` 37.315 lakh crores in 2008–09.
∑ Turnover of Index Futures has recorded a notable increase as the trading
volume has risen from ` 2.37 thousand crores in 2000–01 to ` 3570.11
thousand crores in 2008–09.
∑ In India, Index Options have been more popular in last four years than
the options on individual stock, as more than 65% of the total volume
in options segment has consistently been recorded in terms of Index
Options.
The rest of the book is divided into six chapters. Equity derivatives with an
emphasis on Index option have been discussed in Chapter 2. Chapters 3 and 4
discuss two most commonly used conditions on options price. Further, these
conditions have been tested in the context of Indian derivatives market, in
general, and index options, in particular. Chapter 5 and 6 enumerate implied
volatility in detail. The implied volatilities have been extracted from closing
prices of index options by using Black–Scholes (1973) options pricing model.
Chapter 5 examines its behaviour with respect to different maturities. The
forecasting ability of implied volatility to predict futures volatility and its
comparative performance vis-à-vis historical volatility models have been
examined in Chapter 6. In the last chapter (Chapter 7), evidences borne out by
the secondary data analysis have been further corroborated by the opinion of
Trader Member Organisations (brokerage firms) on pricing related issues. For
this purpose, a questionnaire had been administered amongst the brokerage
houses, actively involved in Future and Options segment, based in National
Capital Region and Mumbai.
2 C H A P T E R

Equity Options and Risk

Management

2.1 INTRODUCTION
Equity options serve, amongst other functions, as an important tool for
managing risk for investors dealing with equity shares and/or portfolio
thereof. These are the derivative contracts that give their holders a right but
not an obligation to buy or sell the underlying asset (equity share or index
of such shares) at a predetermined price called strike/exercise price on or up
to a maturity date. The options market is expected to serve as a tool for risk
hedging, price discovery, enhancing liquidity in the market and, therefore,
facilitate better allocation of capital in an economy. The risk hedging function of
options is invariably claimed as the most important amongst all their functions.
This is in view of the fact that the availability of such financial innovations
in a market helps in transferring risk from one party to another (from buyer
of the options to the seller of the option), and therefore, it helps investors to
take risk which is tailored to best suited their risk appetite.
In addition, it is generally believed and has been demonstrated empirically
across the world that the options market leads the underlying’s market, i.e.
the information gets reflected first in the derivatives market, in general, and
options market, in particular, and then spills over to the underlying’s market. It
would be appropriate to note here that the flow of information is bidirectional;
however, the option market is expected to lead the underlying’s market in
an efficient market environment. This function of the options market can be
traced to the fact that it takes an investor considerably less to have the same
exposure through the options market compared to what it takes in the spot
market to have the same exposure in a financial asset. This is borne out by the
fact that options market entails lower transaction costs and provides leveraged
position to the investors. It is for this reason that the information is expected
to get reflected in the options market prior to the spot market.
Another function of the options market is to enhance liquidity of the
underlying’s market. However, such impacts are difficult to generalize and,
 Equity Options and Risk Management ● 13

therefore, are subject to empirical scrutiny. In sum, these functions of the

options market facilitate allocation of the capital to its most productive usage,
which in fact, is needed for an economy to grow at a faster pace.
Based on the underlying asset, equity options are classified in two
categories—index options and stock options—options on individual stocks.

2.1.1 Index Options

Index options are those derivative contracts that have an index as underlying
asset. Naturally, in case of equity options, index options are floated on an index
of equity shares. In general, such options are floated on the leading index/
indices of equity shares in an economy. For example, in Indian derivatives
market, the most popularly known index option is NSE CNX nifty index
options. These options are floated on one of the leading equity indices of the
Indian economy, i.e. NSE CNX nifty. It is an index of 50 most liquid shares
from different sectors of the economy, which are weighted based on their free
floated market capitalization. The weights of different sectors in the index are
chosen in such a way that the index becomes a proxy for economic activity
of the whole economy. Depending upon the demand in an economy, index
options can be floated on a variety of indices. For example, these may be floated
on sector-specific indices like Banking, Telecom, IT, Infrastructure, Oil & gas,
etc. Besides, indices are also constructed for shares based on their market
capitalization, viz. LARGECAP, MIDCAP and SMALLCAP. The definition
of these categories varies from market to market. Further, these categories get
different definition from time to time, depending upon change in market size.
In Indian derivatives market, NSE offers a total number of eight index options,
e.g. NIFTY-leading index of equity shares, /market index, Bank Nifty, CNX
100, CNXIT, etc. Nifty index options are the most traded amongst all types of
exchange-traded options floated in Indian derivatives market. Guidelines for
floating options on various indices are circulated by SEBI from time to time.
Index options are useful for speculating as well as hedging marker exposure.
Particularly, these options are very useful in managing risk of a variety of
funds, as the options are normally floated on leading indices of an economy
and not on such customized funds. Further, these funds are expected to have
very strong association with the leading index since majority of funds are
benchmarked to leading funds of an economy. Therefore, options on such
leading indices become a natural choice for hedging market exposure of such
tailored funds.

2.1.2 Option on Individual Securities

Options on individual stocks, as the name suggests, are options which have
individual shares as underlying asset. Currently, in Indian derivatives market,
NSE offers options on individual stocks on more than 200 shares from different
14 ● Derivative Markets in India

sectors of the economy. The securities are included in or excluded from the
Futures & Options (F&O) segment list based on the different criteria notified
by SEBI from time to time. Some of such criteria include, average daily market
capitalization and average daily traded value of the security for previous six
months, the market-wide position limit in the security, the quarter sigma
values, etc.
Options on individual stocks are used for speculating as well as hedging
purposes. However, the hedging provided by such options is limited
compared with that provided by index options, as these options can be used
to hedge exposure of underlying security or some other security having strong
correlation with the underlying. On the other hand, index options can be used
to hedge exposure of a variety of funds, since options are not available on
such tailored funds. At the same time, these funds are expected to have strong
association with the leading index/indices on which options are normally
floated in an economy.
In the case of index options, it is important to note that these are essentially
settled in cash. The cash settlement in the case of index options happens
because non-deliverable underlying asset (index), and therefore, physical
settlement becomes impossible. In Indian derivatives market, both types of
options, namely index and stock options, are settled in cash only.
Another important feature of Indian options market is that index as well
as stock options are European in nature, i.e. these can be exercised only at
maturity. It is important to note that options on individual stocks trading at
NSE were being offered as American options till October 2010. In October
2010, NSE decided to move to European style options with an expectation to
improve dearth of liquidity in options on individual stocks.
Further, with respect to maturity of options contracts, both types of options
(i.e. index and stock options) have three variants in terms of their maturity
time: Near the month (NTM)—options having less than or equal to 30 days to
maturity; Next the month (NXTM)—options which are having 31–60 days to
maturity and Far the month (FTM)—options which have more than 60 days
to maturity. In addition, March onwards, Long Dated Options (LDO) were
allowed to trade in Indian options market. However, such options could not
attract significant activity due to liquidity and pricing issues.

2.2 TRADING STRATEGIES USING OPTION FOR PROFIT MAKING

(SPECULATION)
Trading in options with intent to earn profits (profit or trading strategies using
options) is basically speculating in the options market unlike hedging wherein
an investor attempts to protect himself against unfavourable movements
in the market. Speculation essentially dwells upon the view(s) of a trader,
 Equity Options and Risk Management ● 15

primarily, on two important market characteristics/parameters, viz. (i) view

on the direction of the market (i.e. bullish or bearish trend will follow) and
(ii) volatility of the market ( whether market is going to experience larger/
smaller moves; in other words, high or low volatility). Predominantly, based
on these two important characteristics, a trader devices strategy which will
be most profitable, given his views on the market. Some select studies based
on these two parameters have been summarized in Table 2.1.
Table 2.1 Select Trading/ Profit Strategies Using Options: A Classification
Based on Market Direction and Volatility
VOLATILITY
Low volatility High volatility
DIRECTION
MARKET

Bullish Bullish call spread Call option

Bearish Bearish put spread Put option
No view on Butterfly spread Straddle
direction/neutral Box Spread

A detailed discussion on trading strategies summarized in Table 2.1 has been

covered in the following section by dividing them into two major categories:
(i) Directional Strategies and (ii) Non-directional or Neutral strategies. Further,
within a given classification (i.e. directional or non-directional), these strategies
have been clubbed based on trader’s view on volatility.
It is important to note that these strategies can be formed with the help
of European as well as American options. However, in the case of American
options, some of the complex strategies, which include more than one option
contract, may deviate substantially from the expected payoffs. It may happen
on account of the possibility that the two legs of the transaction may not
be exercised at the same time. And, therefore, the actual profit (loss) may
differ substantially from the expected profit, even if the traders view on the
market turns out to be correct. Notably, this risk does not emerge in the case
of European options, as there is no possibility of exercising the option before
maturity.

2.2.1 Directional Trading Strategies Using Options

The directional strategies are those strategies wherein a trader has certain view
on the direction of the market in near future. Since the market can either go up
or down, the traders can have their own view on the market. That is, a trader
who thinks that the market is likely to go up in near future is said to have
bullish view on the market. On the contrary, a trader who thinks market will
go down in near future is said to possess a bearish view. Some of the bullish
and bearish strategies clubbed with traders’ view on volatility are discussed
in following subsection.
16 ● Derivative Markets in India

2.2.1.1 Bullish strategies

The strategies that assume upward direction of the market has further been
classified based on expected magnitude of volatility:
(a) Bullish and high volatility—Call option: In this case a trader thinks
that the market will go up and, at the same time, he believes that the
magnitude of fluctuations will be quite high. In other words, the market
will be on the rising trend and is likely to experience larger moves. In
this situation, the trader should go for a long position in call option to
get maximum benefits in case his views on the market turn out to be
correct.

Example 2.1 Suppose a speculator believes that in a month’s time, the market

will experience a rising trend with fair probability of larger moves. To exploit this
opportunity, he takes long position in a European call option, which is scheduled
to expire one-month from now. The option has a strike price of ` 2950 and is
currently available at ` 100. Further, suppose the underlying share is currently
traded at ` 3000. Since he has purchased a call option, he will be benefited by
upside movement in the market which is large enough to surpass the ` 3050
mark. Since he has paid ` 100 upfront as premium for the call, the price has to
advance by ` 50 from its current level to reach an equilibrium position.
As we know, a call option will be exercised when the underlying assets price
is more than that of the strike price. At maturity, the price can take any value,
given the random nature of the underlying asset; some of such values have
been taken to examine a trader’s profit position for different price scenarios.
The calculation of profit under different price scenarios has been reported in
the following table. Further, for better understanding of the profit position from
the strategy , the payoffs from the strategy have been plotted for select possible
prices/ scenarios at maturity.

Underlying’s Pay-off
price at maturity Action Value (Value-call
(ST) premium)
2900 Abandon 0 –100
2950 Abandon 0 –100
3000 Exercise 50 –50
3050 Exercise 100 0
3100 Exercise 150 50
3150 Exercise 200 100
3200 Exercise 250 150
3250 Exercise 300 200
3300 Exercise 350 250
3350 Exercise 400 300
 Equity Options and Risk Management ● 17

300

250

200

150

100

50

0
50

00

50

00

50

00

50

00

50

00
– 50
28

29

29

30

30

31

31

32

32

33
– 100

– 150 Underlying’s value at maturity (ST)

From the data, it can be easily inferred that the trader will be benefited only
if the market shows an advancing trend. Further, it should move up by at least
` 50 to break-even. He will start generating profit only if the market moves
beyond ` 3050. It is evident from the data that the trader will be benefited only
by large moves; for example, if market reaches at ` 3350 mark. In this scenario,
he will make a profit of ` 250. It becomes a possibility only if the market goes up
by ` 350, which is equivalent to nearly 40% {(350/3000) ¥ 12 } annualized
volatility. In this strategy, the maximum loss that a trader is going to incur is
the premium amount he paid initially; however, profit potential is theoretically
unlimited.

(b) Bullish and low volatility—Bullish call spread: A bullish call spread is
an appropriate trading strategy when a trader expects market to go up
but he expects volatility to be low. It combines two call options, which
are similar in all the aspects (maturity, underlying asset, etc.) except strike
prices. Since the trader thinks that market will go up, he takes a long
position in the call option; simultaneously, he takes a shot position in
the other call option (with higher strike price), as he expects that larger
moves are unlikely. Therefore, he can reduce his initial investment
by selling a call option with higher strike price. In sum, this strategy
involves a long position in a call option with the lower strike price and
simultaneously a short position in a call option with higher strike price.

Example 2.2 Suppose a trader expects that, one month from now the market
will be on the rising trend but large moves are unlikely. To exploit this scenario,
he goes long in a bullish call spread. For the purpose, he takes long position
in a call option having maturity of one month and strike price of ` 3000 (X1),
18 ● Derivative Markets in India

available at ` 60 (C1). At the same time, he takes a short position in another call
option with the same characteristics except the strike price, which he chooses at
` 3070 (X2), currently traded at ` 55 (C2).

Underlying’s Value
Pay-off {Total
price at maturity Bullish call spread
value – (C1 – C2)}
(ST) X1 (Long) X2 (Short) Total
2980 0 0 0 −5
2990 0 0 0 −5
3000 0 0 0 −5
3010 10 0 10 5
3020 20 0 20 15
3030 30 0 30 25
3040 40 0 40 35
3050 50 0 50 45
3060 60 0 60 55
3070 70 0 70 65
3080 80 10 70 65

70

60

50

40

30

20

10

0
2980

2990

3000

3010

3020

3030

3040

3050

3060

3070

3080

3090

– 10
Underlying’s value of maturity (ST)

This strategy results in an initial cash outlay of ` 5, i.e. call premium paid
(on the option the trader is long) net of the premium received from the short
position (C1 – C2). Further, suppose the underlying share/index is currently
traded at ` 3000.
From the data it is evident that the trader will start generating profit only if
share price moves beyond ` 3005. Once the price reaches at ` 3005, the trader
will breakeven. Beyond this point, every increase will add to the profit of the
trader until the price reaches ` 3070. As the price advances beyond this point,
 Equity Options and Risk Management ● 19

the call option you are short in will also be exercised against you and your profit
will stabilize at ` 65 (that is, ` 3070 – 3000 - 5). That is, this strategy will result
to a maximum profit of {X1 – X2 – (C1 – C2)}.
From the data it may be inferred that the trader will be benefited by adopting
this strategy if the market shows bullish trend and, at the same time, shows mild
moves. In case price moves beyond ` 3070, the trader is no longer benefited as
his profit stabilizes at rupee 65. Therefore, such a strategy is the most suitable
one when the market is expected to advance, at the same time large moves are
unlikely. In this systematically the maximum loss that he is going to incur is
the difference between call premiums, i.e. (C1 – C2). This strategy reduces the
maximum loss as well as the maximum profit compared with call option. In case
the market is expected to advance along with larger moves, a long position in call
option becomes more profitable strategy compared with a bullish call spread.

2.2.1.2 Bearish strategies

(a) Bearish and high volatility—Put option: In case a trader believes that
market will experience bearish trend with fair probability of larger
moves (i.e. high volatility), or in other words, the market will be on the
declining trend and is likely to experience larger moves; a long position
in put option becomes most suitable strategy.

Example 2.3 Suppose a trader expects bearish trend in the market clubbed

with larger moves in next one month’s time. To exploit this opportunity, he takes
long position in a European put option, which is scheduled to expire one-month
from now. The option has a strike price of ` 2950 and is currently available at `
50. Further, suppose the underlying share is currently traded at ` 3000. Since
he has purchased the put option, he will be benefited by a downside movement
in the market that is large enough to surpass the 2900 mark. Since he has paid
` 50 upfront as put premium, the price has to go down by more than hundred
points from its current level to fetch an equilibrium position.
We are aware that a put option will be exercised when the underlying assets
price is lower than that of the strike price. At maturity, the price can attain any
value; some of such values have been taken to examine the traders profit position
on their different price scenarios. The calculation of profit payoff of the trader
under different price scenarios has been summarized in the following table.
Further, for better understanding of the position from the strategy, the payoffs
from the strategy have been plotted for different possible prices at maturity.
It is evident from the data that the trader will be benefited only if the market
shows a declining trend. Further, it should move down by at least ` 100 to
bring him to no profit no loss situation. He will start generating profit only if
the market moves below by more than ` 100. From the data it is clear that the
trader will be benefited only by large moves; for example, market touches ` 2650
20 ● Derivative Markets in India

Underlying’s Pay-off
price at maturity Action Value (Value-put
(ST) premium)
2650 Exercise 300 250
2700 Exercise 250 200
2750 Exercise 200 150
2800 Exercise 150 100
2850 Exercise 100 50
2900 Exercise 50 0
2950 Abandon 0 −50
3000 Abandon 0 −50
3050 Abandon 0 −50
3100 Abandon 0 −50

300
250
200
150
100
50
0
2650

2700

2750

2800

2850

2900

2950

3000

3050

3100

– 50
– 100
Underlying’s value at maturity (ST)

mark. In this situation, he will end up with a profit of ` 250. It is possible only
if the market goes down by ` 350, which is equivalent to nearly 40% annualized
volatility. In this strategy, the maximum loss that a trader is going to incur is
the premium amount he paid initially; however, profit potential can theoretically
reach up to the strike price.

(b) Bearish and low volatility—Bearish put spread: A Bearish Put spread
is an appropriate trading strategy when a trader expects market to
decline; however, larger moves are unlikely. A Bearish Put spread
combines two put options, which are similar in all the aspects (maturity,
underlying asset etc.) except strike prices. Since the trader speculates that
market will go down, he takes a long position in put option to benefit
from down movements; simultaneously, he takes a shot position in the
other put option (with lower strike price) as he thinks that larger moves
are unlikely. As a result, he reduces his initial investment by selling
a put option with lower strike price. In sum, this strategy involves
a long position in one put options with the higher strike price and
simultaneously a short position in a put option with lower strike price.
 Equity Options and Risk Management ● 21

Example 2.4 Suppose a trader speculates that one month from now, the market
will be on the declining trend but the volatility will remain low. A bearish
put spread will be an appropriate strategy to get maximum benefit in case the
trader’s expectation turns correct. For the purpose, he takes long position in
a put option having maturity of one month and strike price of ` 3000 (X2),
available at ` 55 (P2). Simultaneously, he goes short in another put option with
the same characteristics but lower strike price, which he chooses at ` 2930 (X1),
currently traded at ` 50 (P1).

Value
Underlying’s price Pay-off {Total
Bearish put spread
at maturity (ST) value-(P2 – P1)}
X1 (Short) X2 (Long) Total
2910 20 90 70 65
2920 10 80 70 65
2930 0 70 70 65
2940 0 60 60 55
2950 0 50 50 45
2960 0 40 40 35
2970 0 30 30 25
2980 0 20 20 15
2990 0 10 10 5
3000 0 0 0 −5
3010 0 0 0 −5
3020 0 0 0 −5

70

60

50

40

30

20

10

0
2910

2920

2930

2940

2950

2960

2970

2980

2990

3000

3010

3020

– 10
Underlying’s value of maturity (ST)
22 ● Derivative Markets in India

This strategy entails an initial cash outlay of ` 5, i.e. put premium paid (on
the option having higher strike price) net of the premium received from the
short position (P2 – P1). Further, suppose the underlying share is currently
traded at ` 3000.
From the data, it is evident that the trader will start earning profit in case
the share price moves below ` 2995. Once the price hits ` 2995 mark, the trader
will breakeven. Beyond ` 2995 mark, every decrease will add to the profit of
the trader until the price reaches ` 2930. As the price declines beyond ` 2930,
the other put option (that you sold) will also be exercised against you and your
profit will stabilize at ` 65 (that is, ` 3000 – 2970 – 5). That is, in this strategy,
maximum profit will be {X2 – X1 – (P2 – P1)}.
From the example, it may be deciphered that this strategy will be appropriate
if the market is likely to decline and, at the same time, larger moves are unlikely.
In case price moves below ` 2930, the trader is no longer benefited as his profit
remains ` 65 beyond this point. Thus, such an strategy is the most suitable
one when the market is expected to decline; however, is likely to exhibit lower
volatility. In a bearish put spread, the maximum loss a trader is expected to
incur is the difference between put premiums, i.e. (P2 – P1). In sum, this strategy
reduces the maximum loss as well as the maximum profit compared with put
option.
In case the market is expected to experience larger moves along with the
bearish sentiment, a long position in put option will be more profitable compared
with a bearish put spread.

2.2.2 Non-directional or Neutral Trading Strategies Using Options

Contrary to directional strategies, Neutral strategies assume that the trade
is neither bullish nor bearish on the market direction. That is, the speculator
thinks that the market can move to either side. A variety of neutral strategies
have been developed in an attempt to maximize profit from the scenarios
where market direction is unpredictable. However, such strategies have
been classified based on trader’s view on volatility of the market—low, high
and neutral volatility. Based on expectation of volatility in the market, some
of popular neutral trading strategies are Straddle, Butterfly Spread and Box
Spread strategy.
(a) Neutral on direction with high volatility—Straddle: Straddle is a
strategy wherein the trader does not assume any direction of the market;
however, he thinks that the larger movers are quite likely. That is, the
market is likely to swing heavily to either direction. In view of this, a
straddle attempts to generate profits in case market experiences larger
moves to either direction. For the purpose, the trader has to take long
position in a call, as well as put option with the same characteristics. A
 Equity Options and Risk Management ● 23

call will ensure maximum profit in case the market turns bullish and
experience larger moves; on the contrary, if market declines and exhibits
high volatility, a long position in put will maximize profit for the trader. It
is important to note that the trader will be able to maximize his profit only
if the market experiences large moves irrespective of their direction.

Example 2.5 Suppose a trader has no view on the direction of the market, i.e.
he fears that the market can go either way in a month’s time; however, he believes
that the volatility is going to be high. To generate profit in this scenario, he takes
a long position in a straddle. For the purpose, he goes long in a call as well as
a put option having strike price of ` 3000 and scheduled to expire one month
from now. The call and put options are currently available at ` 100 and 90,
respectively. Further, suppose the current market price of the underlying share
is ` 3000.

Value
Underlying’s price Pay-off {Total
Straddle Total
at maturity (ST) value − (C + P)}
XC XP value
2500 0 500 500 310
2600 0 400 400 210
2700 0 300 300 110
2800 0 200 200 10
2900 0 100 100 −90
3000 0 0 0 −190
3100 100 0 100 −90
3200 200 0 200 10
3300 300 0 300 110
3400 400 0 400 210
3500 500 0 500 310

400
300
200
100
0
– 100 2500 2600 2700 2800 2900 3000 3100 3200 3300 3400 3500

– 200
Underlying’s value at maturity (ST)
– 300

In the above case, it is evident that the trader will start earning profit if the
market moves up/ down by more than ` 190 from its current level. As long as
market rages with ` 2810–3190, he will incur losses. The trader will breakeven
24 ● Derivative Markets in India

in case the market reaches either of the following two points: ` 2810 and 3190.
The moment market crosses this range to either side, it will start generating
profit to the trader. Further, it is important to note that the trader will be able
to maximize his profit in case the market experiences high volatility. The higher
it move from its current level, higher will the profit that he earns. For example,
the market touches either ` 2500 or 3500 mark, which is possible in case market
exhibits more than 50% volatility, the trader will earn good amount of profit,
i.e. ` 210.
Since this strategy attempts to exploit both the possible directions, it naturally
will cost higher to the trader compared with directional strategies. In this
strategy, maximum loss that a trader can incur will be equal to ` 190 (C + P)
in case market remains at the same level at maturity. In case the market is likely
to experience moderate moves, it is not a suitable strategy. A suitable strategy
for mild moves/low volatility has been discussed next.

(b) Neutral on direction with low volatility—Butterfly spread: A butterfly

spread is an appropriate strategy when a trader has no clue about the
market direction; however, he believes that larger moves are unlikely.
This strategy involves long position in two call options with different
strike prices and short position in the same two call options with the
only difference that the strike for short positions is fixed somewhere
in between the long positions’ strike prices. In doing so, this strategy
results in reduced initial outlay, as a trader takes short position as well.
However, the profit potential promised by the strategy remains limited.
This strategy works well as long as the market swings on either direction
within a moderate limit.

Example 2.6 Suppose a trader who has no idea where the market can move
a month from now is sure that large moves are unlikely. Based on his view, he
takes long position in butterfly spread strategy. For this purpose, he goes long in
two call options—scheduled to expire a month from now—having strike prices
of ` 2950 and 3050, respectively. The options are currently traded at ` 110 and
95, respectively. A the same time, he takes short position in two call option with
the same characteristics with the only difference that the strike price which he
chooses is at ` 3000. This option is currently traded at ` 100. Further, suppose
the underlying share is currently traded at ` 3000. This strategy will result in
an initial cash outlay that is considerably less compared with straddle; at the
same time, profit will also be limited, as it can go up to ` 45, whereas in the case
of straddle it can be substantially high.
From the data, it is evident that this strategy will generate profits as long as
the price remains within moderate limit (exhibits low volatility) irrespective of
direction it takes. It can be easily deciphered from the data that the trader will
 Equity Options and Risk Management ● 25

remain in profitable position in case prices rises/declines by ` 50 from its current

level. Beyond this level, any further movement in the same direction (that is, the
price further declines once it hits ` 2955 mark; or on the other side, it further
advances once it reaches at rupees 3045) will result into loss.

Value
Underlying’s
Butterfly spread Total
price at maturity Total
X1 2*X2 X3 pay-off
(ST) value
(Long) (Short) (Long)
2930 0 0 0 0 −5
2940 0 0 0 0 −5
2950 0 0 0 0 −5
2960 10 0 0 10 5
2970 20 0 0 20 15
2980 30 0 0 30 25
2990 40 0 0 40 35
3000 50 0 0 50 45
3010 60 20 0 40 35
3020 70 40 0 30 25
3030 80 60 0 20 15
3040 90 80 0 10 5
3050 100 100 0 0 −5
3060 110 120 10 0 −5
3070 120 140 20 0 −5

50
40
30
20
10
0
– 10
30
40
50
60
70
80
90
00
10
20
30
40
50
60
70
29
29
29
29
29
29
29
30
30
30
30
30
30
30
30

Underlying’s value at maturity (ST)

From the example, it is clear that the strategy generates maximum profit
when the price remains at the same level, i.e., ` 3000. Further it remains
profitable as long as price moves up or down by less than ` 45. In other word,
this strategy ensures profit in case the market exhibits low volatility irrespective
of its direction.

(c) Neutral on both direction and volatility—Box spread: Box spread is

a suitable strategy when an investor has no clue about the direction
26 ● Derivative Markets in India

as well as the volatility that market might experience in a given span

of time. This strategy is basically a combination of bullish call spread
and bearish put spread. This combination is unique in the sense that it
provides uniform values irrespective of price that the underlying share
takes at maturity. This strategy basically becomes a risk-free strategy
as its profit always remain at a constant level; at the same time, it never
results in loss.

Example 2.7 Suppose a speculator fears that the market can move to either
direction and can experience large as well as small/moderate moves. Based on his
view on the market, he goes long in a box-spread strategy. For this purpose, he
goes long in a bullish call spread as well as in a bearish call spread. For bullish
call spread, consider the same data as discussed in Example 2.2. Further, assume
that a bearish put spread has been constructed with the same strike prices for the
same maturity dates. Assume that such put options with strike prices of ` 3000
and 3070 were traded at ` 58 and 68, respectively. This strategy results in an
initial cash outlay that equals to outlays of the bullish call spread and bearish
put spread added together, i.e. (C1 – C2) + (P2 – P1). It will be ` 15 {(60 − 55)
+ (68 − 58)} in the above mentioned case.
Values of the box-spread strategy for different possible prices/values of the
underlying asset at maturity has been summarized in the following Table. For
better understanding of the behaviour of values from the strategy, these have
been portrayed across various price levels in the figure.

Underlying’s Value
price at Bullish call spread Bearish put spread Box
maturity spread
X1 X2 Total X1 X2 Total
(ST)
Long (Short) (Short) Long
2970 0 0 0 30 100 70 70
2980 0 0 0 20 90 70 70
2990 0 0 0 10 80 70 70
3000 0 0 0 0 70 70 70
3010 10 0 10 0 60 60 70
3020 20 0 20 0 50 50 70
3030 30 0 30 0 40 40 70
3040 40 0 40 0 30 30 70
3050 50 0 50 0 20 20 70
3060 60 0 60 0 10 10 70
3070 70 0 70 0 0 0 70
3080 80 10 70 0 0 0 70
3090 90 20 70 0 0 0 70
3100 100 30 70 0 0 0 70
3110 110 40 70 0 0 0 70
 Equity Options and Risk Management ● 27

Value of box spread strategy

Value

Underlying’s value at maturity (ST)

From the data, it is clear that the combination of a bullish call spread and a
bearish call spread leads to the same value irrespective of the price underlying
asset takes. For this strategy, it is important to note that a trader should go
long in case the value of from the strategy (` 70 in the above-mentioned case)
is more that the initial cash outlay (` 15); otherwise, the trader should shorten
the box spread.
In all the profit strategies discussed, it is very important to note that it would
be safer to go with the European options (especially for combinational strategies),
as in the case of American options, the possibility of early exercise may distort
the expected profit pay-offs.

2.3 USE OF OPTIONS FOR HEDGING RISK

Equity options are extensively used for hedging financial risk/exposer. In fact,
options are introduced with intent to provide suitable risk hedging facility to
the participants in a financial market. Amongst equity options, index options
are of great use to the portfolio managers for managing risk of a variety
of equity funds. An investor who holds a long (short) position in equity
share/portfolio thereof can hedge his risk by taking short (long) position in
appropriate option contract (option on particular equity share/index option)
to hedge against adverse movements in the market.
By virtue of risk hedging facility, options market is claimed to have an
increasing effect on the liquidity of the underlying’s market, as it helps Portfolio
Management Services (PMS) to offer more sophisticated structured/tailor-
made financial products to a variety of investors and, therefore, facilitates
in increasing the mobilization of funds in the capital market. Moreover, the
risk hedging facility provided by the options market motivates risk-averse
investors to invest in the capital market.

2.3.1 Hedging with Index Options

Availability of options on the major indices of the economy helps different
institutional investors (especially, mutual fund organization) to ensure stable
28 ● Derivative Markets in India

earnings to their customers. The derivative products based on the indices of

an economy, e.g. index options, provide an ideal hedging facility to the fund
managers to hedge their market risk, as most of the equity portfolios are
constructed based on the concept of indexing. That is, in some way or other,
their portfolios are the manifestations of the leading index of the country. In
other words, the returns of these funds, in general, show a high correlation with
those of leading indices that makes these derivative instruments on leading
indices an ideal choice for the fund managers to hedge against the market risk.
And therefore, the availability of such financial innovations in the market helps
in restoring confidence amongst the investors even in tough times, which in
turn, facilitates movement of capital and liquidity in the capital market.
Therefore, these innovations help different types of investors to take a
position in the market, depending upon their risk appetite. That is, an investor
can use these instruments to increase or decrease the otherwise unavoidable
risk (systematic risk or beta) of their portfolios, depending upon his choice to
take higher or lower risk.

2.3.2 Changes in the Volume in Options vis-à-vis Futures Considering

the Economic Environment: A Paradigm Shift
The data summarized for the year 2008–09 in Table 2.1 have shown a dramatic
change in the typical pattern of trading in the Indian derivatives market. It is
evident from the fact that the trading volume pertaining to index futures has
been lower compared with that of index options in the year 2008–09. Further,
it may be noted that the trading volume of index futures has consistently been
substantially higher compared to that of index option for the first six years
of its existence. In other words, the index futures market continued with its
typical trend (dominated) in F&O segment from the year 2001–02 to 2007–08;
however, the year 2008–09 has experienced a noticeable shift in the trading
volumes from index futures to index options market. This paradigm shift in
the Indian derivatives market has put this market at par with the developed
markets, which prefer options contracts to futures.
A further examination of the development has been carried out to
understand the reason of this development. For this purpose, the monthly data
on index futures and index options has been collected for the period from April
2008 to June 2009. The data have been summarized in Table 2.1. It is clear for
the first two months of trading, index futures market has clearly dominated
the index options market. However, the index options market started picking
up in volume from June 2008 onwards and has surpassed the index futures
market in the month August 2008. Moreover, the index options market has
consistently been dominating the index futures market thereafter.
 Equity Options and Risk Management ● 29

Table 2.1 Business Growth in Derivatives (F&O) Segment, April 2008 to March 2009 (in crores)
Total
Notional
Turnover of turnover of Share of index Share of index
Month turnover of
index futures derivatives futures options
index options
market
8-Apr 280100 133565 766431 37 17
8-May 267641 129067 797908 34 16
8-Jun 377939 308709 1084064 35 28
8-Jul 395380 357209 1160174 34 31
8-Aug 300449 312102 957445 31 33
8-Sep 380198 461623 1197872 32 39
8-Oct 324962 364510 941646 35 39
8-Nov 256950 292134 745356 34 39
8-Dec 269997 313615 829166 33 38
9-Jan 234141 309271 778118 30 40
9-Feb 205679 305599 712370 29 43
9-Mar 276677 444099 1039930 27 43
Source: www.nse-india.com

Figure 2.1 Relative volume of index options to index futures vis-a-vis total
volume of F & O segment during 2008-09: A paradigm shift

It would be appropriate to note here that the index derivatives (futures and
options) are primarily used for hedging the portfolios against the unfavourable
movement in the market, which may result in substantial reduction in the
value of the portfolio. Such financial innovations are predominantly used by
the fund managers, who are responsible for ensuring stable earnings to their
clients. The use of index derivatives as a hedging tool can be traced to the fact
that most of the funds are created using the concept of indexing. The concept
of indexing connotes benchmarking the portfolio to the best portfolio in the
market, which is the leading index of the economy. Therefore, the derivatives
on such indices are the natural choice for hedging risks that may reduce the
value of the portfolio.
30 ● Derivative Markets in India

In view of the change in the trading pattern of index futures and options
market, it would be appropriate to infer that the investors has shifted their
preferences to the index options for hedging their portfolios. Moreover, the
perceived change in the choice of hedging vehicle can be traced to the dramatic
change in the financial markets across the globe in terms of high volatility of
returns on account of devastative meltdown, which occurred as a result of
the sub-prime crisis. Therefore, it would be appropriate to say that the Indian
investors have shifted from index futures to index options in the wake of high
volatility in the market in view of the fact that the options provide far better
hedging mechanism compared with that of index futures. Major advantage
of hedging through options market (in addition to protection of the portfolio
from an unfavourable movement) is that it allows hedger to take the advantage
of a favourable movement as well.
3 C H A P T E R

Testing Lower Boundary

Conditions for the S&P
CNX Nifty Index Options

3.1 INTRODUCTION
The options markets play a central role in an economy as they enhance better
allocation of capital in securities market by virtue of their functions to facilitate
risk hedging and price discovery. In today’s parlance, where the demand for
the structured finance (which requires excessive use of options contracts) is
booming in India, the role of such markets has acquired greater significance.
The ‘open interests’ in the options segment of Indian derivatives market has
even surpassed that of futures market for last few months since April, 2008.
This development has put the Indian derivatives market at equal footings with
the other international (developed) markets, where the options are preferred
to futures.
There could be two major reasons for such a development. First, increase in
the portfolio management services (PMS), which provide structured financial
products (using options market) to high profile investors. Secondly, an increase
in the variety of the products (in terms of maturity period) has taken place on
account of the introduction of long-dated options on 3rd March, 2008. These
options enable an investor to take a position up to 5 years.
Considering the increasing importance of the options market in India, it
is desired that the market should carry out its required functions in the best
possible way. For the purpose, it is imperative that the market should be
efficient. The reason is that well-functioning financial markets are vital to a
thriving economy, as these markets facilitate price discovery, risk hedging
and allocation of capital to its most productive uses. Inefficiency of a financial
market (e.g. options market in this study) indicates that it is not performing
the best possible job at above-mentioned important functions (Ackert and
Tian, 2000).
The present study attempts to assess the pricing efficiency of the index
options in India using both the spot prices of the underlying asset (i.e. the
daily closing value of the S&P CNX Nifty index) and the futures prices, which
32 ● Derivative Markets in India

are traded on the same underlying asset, the S&P CNX Nifty index. The use
of futures market has been proposed in view of the fact that (i) it helps, to a
marked extent, in ensuring the exploitability of arbitrage opportunities when
underlying asset is an index; (ii) the use of futures markets helps in doing
away with the short-selling constraint as a futures can easily be shorted and
(iii) it costs an investor less to exploit the arbitrage opportunities through
futures market because of the lower transaction costs attached to it and the
leverage they provide.
Notably, the use of futures prices on the same underlying asset instead of
spot prices essentially makes this approach a test of joint market efficiency, as
opined by Fung et al. (1997). At the same time, use of the futures prices facilitates
in assessing the degree of integration or pricing interrelationships between the
different derivative instruments being traded in the financial market (Lee and
Nayar, 1993). In other words, this approach helps in addressing the question
whether market participants consider important pricing interrelationships
while pricing the index options. The scope of the present study is confined to
the pricing interrelation between index options and index futures. In sum, the
use of futures market has been proposed in order to examine the role of futures
market in the absence of short-selling facility in the underlying’s cash market.
In other words, whether the futures market could work as an equally good
alternative to short-selling facility and, therefore, help in restoring equilibrium
in the options market even in the absence of short-selling facility.
The use of futures prices, however, puts one restriction on the otherwise
model-free approach, i.e. it assumes cost-of-carry model to hold. Therefore,
this approach cannot be designated as ‘model-free’ unlike the test of the
boundary condition using spot prices. However, the approach still remains
less restrictive compared with those based on certain pricing models, e.g.
Black and Scholes (1973), which assumes that the stock price and volatility
are governed by some stochastic processes.
In the chapter, the violations or mispricing signals observed from the test
procedures using spot values have been examined as per the specified levels
of liquidity and maturity of options. Also, the violations classified as per the
specified levels of maturity have further been sub-classified according to the
three specified levels of liquidity. The classification facilitates a meaningful
explanation to the exploitability of such violations and, therefore, is very crucial
in assessing the efficiency of the market. This has been done in view of the
fact that mere presence of violations does not indicate market inefficiency; it
is the unexploitability and persistence of such violations which pose serious
concerns/threats to the market efficiency.
Moreover, the learning behaviour of the investors in options markets has
also been examined. This has been done by analysing the number of violations
(from the test of LBCs using spot values) vis-à-vis the number of observations
 Testing Lower Boundary Conditions for the S&P CNX Nifty Index Options ● 33

analysed over the years under reference for both the call and put options. The
learning hypothesis, which requires that the number of violations should go
down over the years, has been proposed to gauge the developments related
to the efficiency of the market. The analysis of violations over the years under
reference is in line with Mittnik and Rieken (2000a), a study in the context of
German stock index options market.
The Section 3.2 of this chapter discusses lower boundary condition using
spot as well as futures prices on the same index that has been tested for options
contracts. The data have been discussed in Section 3.3. Section 3.4 presents
analysis and results. The chapter ends with the concluding observations in
Section 3.5.

3.2 THE LOWER BOUNDARY CONDITIONS

The lower boundary condition, first proposed by Merton (1973a) and further
extended by Galai (1978), plays a crucial role in assessing the options market
efficiency. A number of research studies have been carried out in different
options markets using the lower boundary condition to assess the efficiency
of the markets, including the first one by Galai (1978). The other studies which
tried to diagnose the options market efficiency based on the violation of lower
boundary condition include Bhattacharya (1983), Halpern and Turnbull (1985),
Shastri and Tandon (1985), Chance (1988), Puttonen (1993), Berg et al. (1996),
Akert and Tian (2001), Mittnik and Rieken (2000a).
The lower boundary condition of option prices denotes the minimum price
of an options contract at a given point of time during the life of that options
contract. The violation of the condition indicates arbitrage opportunities.
Therefore, the price for an options contract should necessarily be equal to
or higher than that suggested by the lower boundary condition. In order to
ensure correct pricing in an options market, this is a necessary condition that
needs to be satisfied to uphold the well-known no-arbitrage argument of
options pricing. In literature, the lower boundary condition has been defined
for the European options as well as American options. In this study, as we are
analysing the S&P CNX Nifty index options, which are European (that can
be exercised only at maturity) in nature, the condition defined for European
options constitutes the basis of the study.

3.2.1 The Lower Boundary Conditions Using Spot Values

The lower boundary conditions defined for the call and put options are given
in the Eqs (3.1) and (3.2), respectively, which need to be satisfied in an efficient
market.
ct ≥ max [0, {It – Ke– r(T – t) – TTCt}] (3.1)
pt ≥ max [0, {(Ke– r(T – t) – Tt) – TTCt}] (3.2)
34 ● Derivative Markets in India

In the above equations,

ct is the market price of a call option at time t,
pt is the market price of a put option at time t,
It is the level of underlying index (S&P CNX Nifty) at time t,
K is the strike price of the option contract,
is the expiration time of the option at the time when it was floated,
r is the continuously compounded annual risk-free rate of return,
TTCt is the total transaction costs (i.e. transaction costs relating to trading
in options and spot market) at time t and
(T − t) is the time to maturity of the option at time t (measured in years).
The Eqs (3.1) and (3.2) describe the lower boundary conditions where the
underlying asset is not expected to pay any dividends during the life of the
option. Since, in general, almost all the financial assets pay dividends, the Eqs
(3.1) and (3.2) need to be modified by incorporating dividends. The treatment of
dividends in the test varies based on the assumption made about the payment
of dividends. Some of the studies treated it as a discrete payment, e.g. Smith
(1976), and others as a continuous yield, e.g. Chance (1988). In the present
study, since S&P CNX Nifty index (which includes 50 scrips) based options
are being analysed, it would be difficult to test the lower boundary condition
assuming discrete dividends. Therefore, following Chance (1988), it has been
assumed that dividends are paid as continuously compounded yield. The
lower boundary condition equations for call and put options, assuming that
the index is paying continuously compounded annual dividend yield (d), are
given in the Eqs (3.3) and (3.4), respectively.
ct ≥ max [0, {(e– d (T – t) It – Ke– r(T – t)) – TCCt}] (3.3)
– r(T – t) – d (T – t)
pt ≥ max [0, {( –e It) – TTCt}] (3.4)
The testable forms of the Eqs (3.3) and (3.4), which have been used in the
present study to assess the efficiency of the options market, are given in the
next sub-section.

3.2.1.1 Testable form of the lower boundary conditions

using spot values
Equations (3.3) and (3.4) given above have been rearranged in order to make
them testable to gauge the efficiency of the options markets. The testable
form, to test the Efficient Market Hypothesis (EMH) in terms of lower
boundary condition, is given in the Eqs (3.5) and (3.6) for call and put option,
respectively.
– d (T – t )
ect = [{( e It – Ke r (T t ) ) – TTCt } – ct ] (3.5)
– –

– r (T – t ) – d (T – t )
ept
= [{( Ke –e It ) – TTCt } – pt ] (3.6)
c p
In the above equations, e t and e t are the absolute amount of abnormal
profits (ex-post) or mispricing signals from call and put options, respectively,
 Testing Lower Boundary Conditions for the S&P CNX Nifty Index Options ● 35

if the violation of lower boundary condition takes place. A violation of the

lower boundary condition is recorded if e tc > 0 and e tp > 0 for call and put
options, respectively.

Example 3.1 Exploiting arbitrage opportunities indicated by violation of

Lower Boundary Condition using the underlying’s spot market: A case of Call
Option
On November 10, 2010, a call option contract on S&P CNX Nifty index,
with strike price of ` 6200 and scheduled to expire on November 25, 2010, is
available at ` 90. In spot market, the index is currently being traded at ` 6290.
Suppose that continuously compounded risk-free rate of return is 7.5% p.a.,
and each transaction in F&O segment is subject to transaction cost of 0.05% on
notional value of the contract, i.e. (strike price + premium) ¥ size of the contract.
An option contract on Nifty includes 50 nifty.
Solution As mentioned in the discussion on lower boundary condition, the
minimum price of a call option contract at any given point in time should be
ct ≥ max [{(It – K– r (T – t)) – TCCt}, 0]
Therefore, the minimum price for the above-mentioned call option should be
= Max [{(6290 − 6200 ¥ e−(0.075 ¥ 16/365))
– (0.0005 ¥ (6200 + 90))}, 0] = ` 107.21*
Since the price quoted in market (` 90) is lower than that suggested by lower
boundary condition (under-valuation of the call option), it indicates an arbitrage
opportunity. In case such an opportunity appears, following steps can be taken
to ensure arbitrage gains.
Steps required now (on spot):
All steps required now need to be taken simultaneously to lock-in arbitrage
profit. This will hold true for Examples 3.1–3.3.
Step 1: Purchase the call option as it is undervalued.
Step 2: Short-sell the underlying asset (index) at current market price.
Step 3: Invest the remaining amount ` 6,196.85 (` 6290Spot value – ` 90Call premium
– ` 3.15Transaction cost) at risk-free rate for the remaining life of the contract.
Steps required at maturity:
Step 4: Liquidate your investment in the risk-free asset. You will realize a sum
of ` 6196.85 ¥ e(0.075 ¥ 16/365); that is, ` 6217.26.
Step 5: Square off your short position in the underlying asset. For this, (a)
exercise the option contract and buy it at ` 6200 if the underlying asset is being
traded at more than ` 6200. (b) On the contrary, if the underlying is traded at
*
It is important to note that dividend yield and transaction cost related to the spot market have
been ignored in determining minimum price of the option contract in order to avoid further
complexity.
36 ● Derivative Markets in India

a lower price, e.g. ` 6150, abandon the option to purchase the underlying asset
at ` 6200, and purchase it directly from the spot market.
From the above, it is clear that, in any case, maximum amount required to square
off the short position is ` 6,200. Since you received ` 6,217.26 from your initial
investment, you end up generating a profit of ` 17.26 (` 6,217.26 − 6,200).
And, on the whole contract (which includes 50 Nifty), you will end up with
the profit of ` 863.

Example 3.2 Exploiting arbitrage opportunities indicated by violation of

Lower Boundary Condition using the underlying’s spot market: A case of put
option
On November 10, 2010, a put option contract on S&P CNX Nifty index, with
strike price of ` 6200 and scheduled to expire on November 25, 2010, is currently
available at ` 70. In the spot market, the index is currently being traded at
` 6090. Suppose that continuously compounded risk-free rate of return is 7.5%
p.a., and each transaction in F&O segment is subject to transaction cost of
0.05% on notional value of the contract.
Solution As mentioned in the discussion on lower boundary condition, the
minimum price of a put option contract at any given point in time should be
pt ≥ max [{(Ke– r (T – t) – It) – TTCt}, 0]
Therefore, the minimum price for the above-mentioned put option should be
= Max [{(e−(0.075 ¥ 16/365) ¥ 6200 − 6090)
– (0.0005 ¥ (6200 + 70))}, 0] = ` 86.52
Since the price quoted in the market (` 70) is lower than that suggested
by lower boundary condition (under valuation of put option), it indicates an
arbitrage opportunity. In case such an opportunity appears, following steps
need to be taken to ensure arbitrage gains.
Steps required now (on spot):
Step 1: Purchase the put option as it is undervalued.
Step 2: Take long position (purchase) in the underlying asset (index) at current
market price.
Step 3: Borrow a sum of ` 6,163.15 (` 6,090Spot value + ` 70Put premium
+ ` 3.15Transaction cost) at risk-free rate for the remaining life of contract.
Steps required at maturity:
Step 4: Square off your long position in the underlying asset. For this, (a) in case
the underlying asset is being traded at less than ` 6200, exercise the put option
and sell the underlying asset at ` 6200; (b) On the contrary, if the underlying is
traded at a higher price, e.g. ` 6250; abandon the option to sell the underlying
asset at ` 6200 and sell it directly in the spot market.
 Testing Lower Boundary Conditions for the S&P CNX Nifty Index Options ● 37

Step 5: Pay off the borrowed sum along with the accrued interest for the period.
The amount payable is ` 6183.45 (` 6163.15 ¥ e(0.075 ¥ 16/365)).
From the above, it is clear that, in any case, you will realize ` 6,200 or more by
selling the underlying asset (squaring off your long position); whereas, amount
needed to pay off the borrowed sum is ` 6183.45. As a result, you will make
a profit of ` 16.55 (` 6,200 – ` 6183.45). And, on the whole contract (which
includes 50 Nifty),; you will book a profit to the tune of ` 827.50.
Note: In the case of index derivatives, it is important to note that final
settlement takes place in terms of cash, as physical settlement (delivery of
the underlying asset) is not possible. The cash settlement is entailed in view
of the nature of underlying asset, i.e. index, and the whole index cannot be
delivered. Therefore, the price differences are settled instead of delivery/receipt
of the underlying. For example, holder of a call option exercises his option
when value of the underlying index is ` 6,300; he will get a difference of ` 100
in case the strike price was ` 6,200. Similarly, in the case of put option with
the same characteristics, the holder will receive a sum of ` 50 on exercising
his option in case the value of underlying index turns out ` 6150.

3.2.2 The Lower Boundary Conditions Using Futures Prices

The test of lower boundary condition using futures prices is in line with
Puttonen (1993)—a study done in the Finnish index options market. Moreover,
the test of options market efficiency using futures prices (on the same
underlying asset) is in line with Lee and Nayar (1993), Fung and Chang (1994),
Fung et al. (1997), Fung and Fung (1997), Fung and Mok (2001), etc. with the
only difference that the condition tested in the study is the lower boundary
condition for the options prices, whereas all above studies focus on the put-
call-futures parity condition.
The lower boundary conditions using corresponding futures prices (with
the same maturity date) are given in the Eqs (3.7) and (3.8), respectively, for
the call and put options. These conditions are expected to hold in an efficient
options market.
ct ≥ max [0, {e– r(T – t) (Ft – K) – TTCt*}] (3.7)
– r(T – t)
pt ≥ max [0, {e (K – Ft) – TTCt*}] (3.8)
In the above equations, Ft is the value of the S&P CNX Nifty futures (with
same expiration date as of the option under consideration) at time t and
TTCt is the total transaction costs (i.e., transaction costs relating to trading in
options and futures contracts) at time t. All other variables are the same as in
Eqs (3.1) and (3.2).
The dividends expected from the underlying asset during the life of the
option have been ignored, since the underlying asset used in the test is futures
prices/values of the index instead of the spot prices/values. This has been done
38 ● Derivative Markets in India

due to the fact that the futures prices (in an efficient market) are expected to
have impounded the effect of dividends on the prices of the underlying asset.
3.2.2.1 Testable form of the lower boundary condition
using futures prices
The Eqs (3.7) and (3.8) have been rearranged in order to make them testable
to gauge the efficiency of the options market. The testable form, to gauge the
EMH using lower boundary condition, is given in the Eqs (3.9) and (3.10) for
call and put option, respectively.
– r (T – t )
e tc = [{( e ( Ft – K )) – TTCt* } – ct ] (3.9)
– r (T – t )
e tp = [{( e ( K – Ft )) – TTCt* } – pt ] (3.10)
c p
In the above equations, e t and e t denote the absolute amount of abnormal
profits (ex-post) or mispricing signals from call and put options, respectively,
if the violation of lower boundary condition occurs. A violation of the lower
boundary condition is recorded if e tc > 0 and e tp > 0 for call and put options,
respectively. Though the presence of such profits is indicative of market
inefficiency, it should not be treated as a conclusive remark on the efficiency
of the market.
It may be noted that all equations relating to test of LBCs using spot as well
as futures prices have been specified considering the transaction costs but
assuming zero or negligible bid-ask spread. In view of this, there is always
a chance that the arbitrage opportunities suggested by these equations may
disappear in the presence of the bid-ask spread, especially, for the options
traded relatively less frequently. Therefore, due consideration has been given
to the bid-ask spreads while interpreting the violations to draw inferences
regarding the market efficiency. The details on the transaction costs included in
the analysis have been summarized in the data section. On the contrary, given
the fact that the bid-ask spread for options is not included in the transaction
database provided by NSE and the difficulty to estimate such costs, it has
been excluded in the above equations. In operational terms, our study is in
line with that of Halpern and Turnbull (1985).
In this regard, commenting upon the exploitability of observed mispricing
signals, Trippi (1977) and Chiras and Manaster (1977) concluded that the
signals so observed were exploitable using a specified trading strategy to
ensure ex-ante exploitation of such profit opportunities. However, in the
present study, no strategy has been executed to ensure ex-ante exploitability
of abnormal profits suggested by mispricing signals, as the test procedure
applied is ex-post in nature.

Example 3.3 Exploiting arbitrage opportunities by using the underlying’s

future market: A case of call option
In addition to the data given in Example 3.1, assume that a futures contract
on NSE CNX Nifty index, scheduled to expire on November 25, 2010, is traded
 Testing Lower Boundary Conditions for the S&P CNX Nifty Index Options ● 39

at ` 6310. Examine the arbitrage opportunity, if any. Determine the profit which
can be generated by exploiting this opportunity.
Solution As mentioned in the discussion on LBC using futures prices, the
minimum price of a call option contract at any given point in time should be
ct ≥ max [{e–r(T – t) (Ft – K) – TTCt*}, 0]
Therefore, the minimum price for the above-mentioned call option should be
= Max [{e− (0.075 ¥ 16/365) ¥ (` 6310 − ` 6200)
– (0.0005 ¥ (6200 + 90 + 6310))}, 0] = ` 103.34*
Since the price quoted in market (` 90) is lower than that suggested by
the boundary condition, it indicates an arbitrage opportunity. In case such
an opportunity appears in the market, the following steps need to be taken to
ensure arbitrage gains.
Steps required now (on Spot):
Step 1: Buy the call option at the current market price as it is undervalued.
Step 2: Short (sell) the future contract at current market price.
Step 3: Borrow ` 96.30 (` 90Call premium + ` 6.30Transaction cost ) at risk-free rate
for the remaining time to maturity.
Steps required at maturity: At maturity, price of the underlying asset will
follow one of the following three scenarios: (i) less than the strike price of the
option contract, (ii) more than the strike price but less than or equal to future
price (Ft) and (iii) more than futures price (Ft).
Scenario 1: Less than the strike price of call option contract, i.e. price of the
underlying index turns out to be less than ` 6,200. For example, at maturity,
the index is traded at ` 6150.
Step 4: You will not exercise the call option as market price of the asset is less
than strike price (` 6200) of the contract.
Step 5: You will make a gain of ` 160 (` 6310 − ` 6150) on short position in
futures. Since you entered into a futures contract to sell the underlying asset
at ` 6310 on the maturity of the contract irrespective of price of the underlying
asset in spot market at maturity and as the current market price in the spot
market turns out to be ` 6150, you will make a gain as you will be able to sell
the asset at ` 6310.
Scenario 2: More than the strike price of call option contract but less than or
equal to the price of the futures contract. For example, at maturity, the index
is traded at ` 6250.
Step 4: You will exercise the call option, as the price of the asset is more than
the strike price (` 6200) of the contract. And, you will gain ` 50 on exercising

*
It may be noted that the margin needed to take a position in futures market and cost thereof
have been ignored in determining minimum price of the option contract in order to avoid further
complexity.
40 ● Derivative Markets in India

the call option, as you have the right to purchase the asset at ` 6200, which is
currently traded at ` 6250.
Step 5: Similarly, you will make a gain of ` 60 on the short position in
futures.
In sum, you will make a total gain of ` 110 from your portfolio (long in call and
short in futures). Further, it may be noted that this portfolio will generate ` 110,
in case price of the underlying asset ranges between ` 6200 and ` 6310.
Scenario 3: More than the price of the futures contract. For example, at
maturity, the index is traded at ` 6400.
Step 4: You will exercise the call option as the price of the asset is more than
the strike price (` 6200) of the contract. And, you will gain ` 200 on exercising
the call option.
Step 5: However, you will lose ` 90 on the short position in futures.
In sum, you will make a total gain of ` 110 from your portfolio (long in call
and short in futures). It is important to note that this portfolio will generate a
profit of ` 110, in case price of the underlying asset turns out to be more than
` 6310.
Finally, from all the three possible scenarios for the price of the underlying asset,
it is evident that an arbitrageur will always earn ` 110 or more.
Step 6: You need to pay off the borrowed sum along with the accrued interest,
i.e. ` 96.62 (` 96.30 ¥ e (0.075 ¥ 16/365)).
From the above, it is clear that although, in any case, you earn ` 110 or more,
you require ` 96.62 to pay off your borrowings. In other words, you make a
minimum profit of ` 13.38 (` 110 − ` 96.62). And, on the whole contract (which
includes 50 Nifty); you will end up with the profit of ` 669.

Example 3.4 Exploiting arbitrage opportunities by using the underlying’s

future market: A case of put option
In addition to the data given in Example 3.2, further assume that a futures
contract on NSE CNX Nifty index, scheduled to expire on November 25, 2010,
is traded at ` 6110. Examine the arbitrage opportunity, if any. Determine the
profit which can be generated by exploiting this opportunity.
Solution As mentioned in the discussion on LBC using futures prices, the
minimum price of a put option at any given point in time should be
pt ≥ max [{e–r(T – t) (K – Ft) – TTCt*}, 0]
Therefore, the minimum price for the above-mentioned call option should be
= Max [{e− (0.075 ¥ 16/365) ¥ (` 6200 − ` 6110)
– (0.0005 ¥ (6200 + 90 + 6110))}, 0] = ` 83.50
Since the price quoted in the market (` 70) is lower than that suggested by
the boundary condition, it indicates an arbitrage opportunity. In case such
 Testing Lower Boundary Conditions for the S&P CNX Nifty Index Options ● 41

an opportunity appears in the market, the following steps need to be taken to

ensure arbitrage gains.
Steps required now (on Spot):
Step 1: Buy the put option at the current market price, as it is undervalued.
Step 2: Take long position in the futures contract at current market price.
Step 3: Borrow ` 76.20 (` 70Put premium + ` 6.20Transaction cost) at risk-free
rate for the remaining time to maturity.
Steps required at maturity: At maturity, price of the underlying asset will
follow one of the following three scenarios: (i) less than the price of the futures
contract (Ft), (ii) more than the futures price (Ft) but less than or equal to strike
price of the put options and (iii) more than the strike price of the options.
Scenario 1: Less than the price of the futures contract. For example, at maturity,
the index is traded at ` 6050.
Step 4: You will exercise the put option, as current price of the asset is less
than the strike price (` 6200) of the contract. And, you will make a gain of `
150 (` 6200 − ` 6050).
Step 5: On the other hand, you will incur a loss of ` 60 (` 6110 – ` 6050) on
long position in futures. Since you entered into a futures contract to purchase
the underlying asset at ` 6110 on maturity of the contract, irrespective of its
price at maturity, and as the current market price in the spot market turns out
to be ` 6050, you incur a loss as you have to buy the asset at ` 6110, which is
currently selling at ` 6050 in the spot market.
In sum, you will make a total gain of ` 90 from your portfolio.
Scenario 2: More than the futures price but less than or equal to the strike
price of the put options contract. For example, at maturity, the index is traded
at ` 6150.
Step 4: You will exercise the put option as the price of the asset is still less than
the strike price (` 6200) of the contract. And, you will gain ` 50 on exercising
the option as you have the right to sell the asset at ` 6200, which is currently
traded at ` 6150.
Step 5: Similarly, you will make a gain of ` 40 on the long position in
futures.
In sum, you will make a total gain of ` 90 from your portfolio.
Further, it is important to note that this portfolio will generate ` 90, in case
price of the underlying asset is less than or equal to ` 6200.
Scenario 3: More than the strike price of the put option. For example, at
maturity, the index is traded at ` 6250.
Step 4: You will not exercise the put option as the price of the asset is more than
the strike price (` 6200) of contract.
Step 5: However, you will make a gain of ` 140 on the long position in
futures.
42 ● Derivative Markets in India

In sum, you will make a total gain of ` 140 from your portfolio.
Finally, from all the three possible scenarios for the price of the underlying asset,
it is evident that an arbitrageur will always earn ` 90 or more.
Step 6: You need to pay back the borrowed sum (along with the accrued interest),
i.e. ` 76.45 (` 76.20 ¥ e (0.075 ¥ 16/365)).
From the above, it is clear that, in any case, you will earn ` 90 or more; whereas,
you require ` 76.45 to pay off your borrowings. In other words, you will make a
minimum profit of ` 13.55 (` 90 – ` 76.45). And, on the whole contract (which
includes 50 Nifty), you will end up with a profit of ` 677.50.
Note: While assessing options market using futures market, it is assumed
that futures market is efficient. That is, it will converge to spot price at the
maturity of the contract.

3.4 THE DATA

This study has attempted to analyse secondary as well as primary data on the
options market. The secondary data have been collected from the websites of
National Stock Exchange of India Ltd. and Reserve Bank of India (RBI). The
primary data have been collected through a survey conducted among the
brokerage firms.

3.4.1 The Secondary Data

3.4.1.1 Data related to options contracts, spot market, futures market

and interest rates
The secondary data considered for the analysis can be broadly classified into
four categories: (i) data related to S&P CNX Nifty index options contracts,
(ii) daily closing values of the S&P CNX Nifty index, (iii) data related to the
futures contracts, i.e. the S&P CNX Nifty index futures, and (iv) data on the
risk-free rate of return.
The first dataset relates to options, it consists of daily closing prices of
options, strike prices, deal dates, maturity dates and number of contracts of
call and put options, respectively. In order to minimize the bias associated
with nonsynchonous tradingi, only liquid option quotationsii are being considered
for the analysis. The next data set consists of the daily closing values of the
S&P CNX Nifty index. The third data set is regarding the futures contracts.
It includes daily closing prices of S&P CNX Nifty index futures, deal dates,
maturity dates and number of contracts traded. The fourth data set constitutes
of monthly average yield on 91-days Treasury-bills. The yield on T-bills has
been converted into continuously compounded annual rate of return using
the relationship given in Eq. (3.1).
 Testing Lower Boundary Conditions for the S&P CNX Nifty Index Options ● 43

r = ln (1 + r*) (3.11)
Where, r is the proxy for continuously compounded annual risk-free rate
of return and r* is the average annual yield on 91-days T-bill of the maturity
corresponding to the maturity date of the options contact.
The data for all the four mentioned categories have been collected from
June 4, 2001 (starting date for index options in Indian securities market), to
June 30, 2007. The first, second and third data sets have been collected from
the website of NSE, and the fourth category of data set has been collected
from website of RBI.

3.4.1.2 Transaction costs

The testing of model-free approach, i.e. Lower Boundary Conditions (LBCs)
and Put-Call Parity (PCP) provides the flexibility of directly incorporating the
transaction costs, which need to be reckoned in order to have true assessment
of market efficiency. In view of this, the transaction cost has been estimated
by interacting with the trading members located at Delhi. Moreover, some of
the Indian studies on the subject, e.g. Vipul (2008), have been referred to in
order to have a justifiable estimate of transaction costs.
The transaction costs typically include brokerage charged by the brokerage
houses/trading members of the exchanges, service tax on the brokerage,
stamp duty, opportunity cost of the margin deposits required in the case of
futures contracts and short options positions, etc. In Indian capital market,
another charge, namely Securities Transactions Tax (STT), was introduced and
implemented with effect from October 1, 2004. Notably, such charges were
to be levied only on the sell side of the transactions in the derivatives market
unlike the equity market transactions, where STT was proposed to be levied on
both legs of the transactions. In view of this, the transaction costs considered
in the study typically include brokerage, service tax on the brokerage and STT
(October 1, 2004, onwards).
Since the transaction costs constitute a major constraint to arbitrage (Ofek
et al., 2004), an attempt has been made to have an estimate of such costs in
Indian derivatives market. For the purpose, interviews were conducted with
the senior employees of brokerage houses based at Delhi, India. A consensus
was arrived at an estimate of brokerage of 0.05% (including service taxes)
for Futures and Options (F&O) in the case of retail investors. Such costs are
charges at ‘(strike price + premium) ¥ lot size’ for options contracts and ‘Futures
price at the time of the transaction ¥ lot size’ for futures contracts. However, it
may be noted that such costs may go down to 0.03% (including service taxes)
for the institutional investors. Besides, another category of investors who are
more likely to get benefited from such arbitrage opportunities, the brokerage
houses, bear the least cost of trading amongst all types of investors/players in
the market, as they are not required to pay any brokerage. However, it would
44 ● Derivative Markets in India

be reasonable to consider the opportunity cost for the brokerage house, and
a logical estimate could be the cost incurred by the institutional investors, i.e.
0.03% in the case of F&O segment, as pointed out by Vipul (2008). Moreover,
with respect to the transaction costs pertaining to spot market transactions for
the trading member organizations, an estimate of 0.15% (excluding STT) of
the traded value has been arrived at, based on the interaction with different
trading member organizations based at Delhi and as suggested by Vipul
(2008). Based on these estimates, the violations have been calculated using
the transaction costs applicable to the trading member organization, given
their least cost structure.
Besides, the STT charge of 0.01% (on the sell side of the transactions in
F&O segment) has also been included in the transaction costs. Likewise, the
STT charge of 0.125% has been considered for the spot market transaction,
notably, for both legs of the transaction. Since the STT was introduced in
October, 2004, it has been considered as part of the transaction costs for the
arbitrage opportunities that occurred after 1st October, 2004. In sum, for the
analysis purpose, the major constituents of the transaction costs have been the
brokerage and the service tax on it before October, 2004; and it additionally
includes STT October 1, 2004, onwards. The definition of the transaction costs
have been confined to the brokerage and STT (wherever applicable) as these
constitute, in general, more than 90% of the transaction costs (excluding bid-
ask spread). Though the analysis has been conducted ignoring the bid-ask
spread and opportunity cost of the margin deposits, these have been given
due consideration while ensuring the exploitability of mispricing signal. This
has been done in view of the fact that bid-ask spread, in particular, plays a
very important role in assessing the options market efficiency, as opined by
Baesel et al. (1983) and Phillips and Smith (1980).

3.5 ANALYSIS AND EMPIRICAL RESULTS

3.5.1 Analysis of Magnitude of the Violations

The ex-post analysis of the call and put options has been carried out on the
basis of the Eqs (3.5) and (3.6) for the period of 6 years starting from June,
2001, to June, 2007. In other word, the analysis of LBCs using spot market
has been the basis of the study. The condition has been tested for 40,298 and
35,171 daily liquid quotes for call and put options, respectively. The results
summarized in Table 3.1 reveal that the total number of violations observed
were 7019 out of total observations of 40,298 in the case of call options that
amounts to 17.42% of total number of observations analysed. Likewise,
1,544 violations were found out of 35171 observations for put options that
accounts for 4.39% of the total number of observations examined. It may be
noted that these violations (i.e. the value of e tc and e tp turn out to be positive)
 Testing Lower Boundary Conditions for the S&P CNX Nifty Index Options ● 45

were recorded when no transaction costs were reckoned. However, these

numbers reduced substantially to 2892 and 707 on account of introduction of
transaction costs. And, therefore, the violations after reckoning the transaction
costs (the additional cost an arbitrageur has to incur while trying to exploit
such opportunities) have been taken up for further analysis in view of their
empirical appeal.
Table 3.1 Violations of the Lower Boundary Condition and Liquidity Levels, June 2001–07
Call options Put options
Particulars
(Percentage) (Percentage)
Total number of observation analysed 40,298 35,171
Total number of violations observed before 7019 1544
transaction costs (17.42) (4.39)
Total number of violations observed a er 2892 707
transaction costs (7.18) (2.01)
Violations relating to the three specified levels of liquidity
2523 654
(a) Thinly traded options
(87.24) (92.50)
278 46
(b) Moderately traded options
(9.61) (6.51)
91 7
(c) Highly traded options
(3.15) (0.99)
Total 2892 707
In the table, thinly, moderately and highly traded options, respectively, denote those
option contracts which have 100, 101–500, more than 500 contracts traded per day.

As far as the frequency of violations is concerned, the results indicate that

it (mispricing signals) is significantly higher for the call options (about 7.18%
of the total observation analysed) compared with those in the case of put
options (about 2.01% of the total observation analysed). Further, in order to
have better insights about the behaviour of the mispricing signals obtained
from the LBC, the violations (after reckoning the transaction costs) have been
examined with respect to liquidity and maturity of the options contracts.
Liquidity of the options has been decomposed into three levels based on the
trading volume: (i) thinly traded options, which have 1–100 contracts traded
per day; (ii) moderately traded options, which have 101–500 contracts traded
per day; and (iii) highly traded options, which have more than 500 contracts
traded per day. Likewise, Maturity of the options has been classified into four
levels: (i) 0-7 days to maturity; (ii) 8-30 days to maturity; (iii) 31-60 days to
maturity; and (iv) 61-90 days to maturity.
Notably, the violations so classified as per the specified levels of maturity
have further been sub-classified as per the three specified levels of liquidity
due to the fact that liquidity constitutes the basis for exploitation of arbitrage
opportunities. The proposed classifications and their sub-classifications
46 ● Derivative Markets in India

facilitate in drawing some meaningful inferences about the exploitability

of the observed mispricing signals, which in turn, help to assess the role of
the existing market-microstructure in restoring market efficiency in Indian
derivative market. The results related to the violations, classified as per
specified levels of liquidity and maturity along with their sub-classifications,
are summarized in Tables 3.2 and 3.3, respectively.
It may be pertinent to note that the results across specified levels of liquidity
have direct implications for the exploitability of the observed mispricing
signals, as the higher the liquidity is, the lower would be the trading cost and
bid-ask spread. Also, higher liquidity ensures execution of the trading strategy
required to tap such abnormal profits. In contrast, the different specified
levels of maturity have an indirect impact since these primarily influence
the liquidity and, hence, the exploitability of such violations. Therefore, the
behaviour of violations with respect to maturity has been interpreted in light
of the liquidity levels corresponding to their specified levels.
The frequency of violations regarding the different levels of liquidity is
summarized in Table 3.1 and the analysis of their magnitudes is reported
in Table 3.2. A vast majority of the violations in this category, i.e. about 87%
and 93% of the violations in the case of call and put options, respectively, are
confined to the thinly traded options, which can be designated as unexploitable
because of (i) higher bid-ask spread and (ii) difficulty in implementation of
the strategy.
Also, it may be noted that the magnitude of violations decreases as the
liquidity increases. Precisely, for the highly liquid contracts (which are
approximately 3 in the case of call options, and a meager 1% in the case of
put options), where the possibility of exploitability is quite high as the bid-ask
spread is expected to be considerably low, the mean of abnormal profits per
lot are merely Rs. 91 for the violations pertaining to call options; however, a
relatively higher magnitude has been recoded in the case of put options, i.e.
Rs. 341. In general, such meagre exploitable profit opportunities are clearly
not attractive propositions for the retail arbitrageurs as they are subject to
higher transaction costs. However, these opportunities might be exploitable for
the institutional investors and the trading member organizations (Brokerage
firms).
Moreover, amongst the violations relating to highly liquid contracts, only
25% observations, i.e. the third quartile (as reported in the Table 3.2), seem
to be exploitable as these offer relatively higher profit of more than 338 and
510 for call and put options, respectively, which is quite likely to be profitable
after reckoning the bid-ask spread. The remaining 75% observations amongst
highly liquid category yield returns that could be designated as fairly below
the exploitable level in the presence of bid-ask spread costs. Similarly, in
 Table 3.2 Liquidity-wise Descriptive Statistics for Violations of the Put-call Parity Condition for Under-priced
and Over-priced Put Options in Indian Securities Market, June 2001–07
Call options Put options
Liquidity Number of violations Magnitude of violations (`) Number of violations Magnitude of violations (`)
(Percentage) Mean S.D. Q1 Q2 Q3 (Percentage) Mean S.D. Q1 Q2 Q3
2523 654
Thinly traded 1165 2576 156 404 963 1375 2966 150 399 1268
(87.24) (92.50)
Moderately 278 46
341 411 83 231 462 1183 3172 81 244 533
traded (9.61) (6.51)
91 7
Highly traded 269 292 70 159 338 341 211 192 281 510
(3.15) (0.99)
Total 2892 1058 2427 139 368 1136 707 1352 2965 145 382 1254
S.D., Q1, Q2 and Q3 denote standard deviation, first quartile, second quartile (i.e. median) and third quartile, respectively.
The values have been rounded off to zero decimal places.
Testing Lower Boundary Conditions for the S&P CNX Nifty Index Options
●
47
48 ● Derivative Markets in India

the case of moderately traded options, the third quartile seems to offers
relatively attractive profits (more than ` 462 and 533 for call and put options,
respectively), and in the presence of bid-ask spread are quite likely to be
exploited. In sum, nearly 3% (25% of 12.76%) of the violations for the call
options seems to be exploitable by trading member organizations. Likewise,
a relatively lower proportion of violations, i.e. less than 2% (25% of 7.50%),
can be designated as exploitable in the case of put options.
In another significant observation, the frequency of violations across
different levels of maturity signifies a decreasing trend with an increase in
time to maturity when the first two levels (0–7 and 8–30 day to maturity)
are clubbed together. The results are reported in Table 3.3. In addition, a
majority of the mispricing signals are confined to the options having 0–7
and 8–30 days to maturity—approximately 83% in call options and 91% in
put options. However, for the next two levels, i.e. 31–60 and 61–90 days to
maturity, the combined percentage is merely 17% and 9% for call and put
options, respectively. The concentration of the violations in the 0–30 days to
maturity category and, especially, in 0–7 days to maturity is similar to that
reported by Bhattacharya (1983), a study in the context of US market where
42% of the total violations had one week or less to maturity.
The concentration of violations in 0–7 days to maturity category (especially
in the case of put options) can be attributed to the fact that most of the
arbitrageurs, in general, try to unwind their arbitrage positions when the
options are nearing maturity. On account of this, the liquidity in such options
is expected to be very thin, as there are only a few or no buyers. This, in turn,
causes the transaction costs, especially the bid-ask spread, to be considerably
high. Therefore, the lack of liquidity and less time to maturity might be cited
as the major reasons for the observed mispricing signal remaining unexploited.
This is eloquently borne out by the fact that the majority of violations in this
category (viz. 87% in the case of call options and 91% in the case of put options)
belong to the thinly traded options category and, therefore, may safely be
designated as unexploitable.
Moreover, if we look at the violations in this category, which had high
traded volume, offer very low profit opportunities as the third quartile start
with profits as low as ` 147 in the case of call options; however, the figure is
relatively attractive, ` 423, in the case of put options but with the very low
frequency. Likewise, the third quartile of such violation having moderately
traded volume offers possibly exploitable opportunities in the light of bid-
ask spread.
The behavior of violations pertaining to the 8–30 days to maturity category
is quite similar to that of 0–7 days to maturity category, as the violations
belonging to the highly liquid category are nearly 3% in the case of call
options and less than 1% in the cases of put options (of the total violations
 Table 3.3 Maturity-wise Descriptive Statistics for Violations of the Put-call Parity Condition for Under-priced
and Over-priced Put options in Indian Securities Market, June 2001–07
Call options Put options
Number of Magnitude of violations (`) Number of Magnitude of violations (`)
Days to Liquidity
maturity violations violations
(Percentage) Mean S.D. Q1 Q2 Q3 (Percentage) Mean S.D. Q1 Q2 Q3
Thinly 617 359
1569 3173 169 437 1326 1222 2542 144 374 1217
traded (87.39) (91.35)
Moderately 66 29
232 277 50 118 321 1491 3847 56 236 562
0–7 traded (9.35) (7.38)
Days Highly 23 5
136 165 50 89 147 341 185 249 281 423
traded (3.26) (1.27)
706 393
Overall 1397 3002 139 369 1143 1231 2640 135 372 1164
(24.41) (55.59)
Thinly 1448 234
990 2117 137 390 815 1632 3579 154 453 1394
traded (86.14) (93.60)
Moderately 177 14
345 348 96 259 462 710 1530 122 256 368
8–30 traded (10.53) (5.60)
Days Highly 56 2
262 259 95 208 336 343 361 87 343 598
traded (3.33) (0.80)
1681 250
Overall 898 1982 131 356 731 1570 3488 151 425 1335
(58.13) (35.36)
Thinly 415 55
1119 2962 193 443 888 1175 2611 151 350 1285
traded (89.83) (94.83)
Moderately 35 3
524 736 138 300 656 421 412 194 317 596
31–60 traded (7.57) (5.17)
Days Highly 12
559 421 257 518 749 Zero NA NA NA NA NA
traded (2.60)
462 58
Testing Lower Boundary Conditions for the S&P CNX Nifty Index Options

Overall 1059 2821 189 440 853 1136 2548 146 349 1265
(15.98) (8.20)
●
49
 50
●

Call options Put options

Derivative Markets in India

Number of Magnitude of violations (`) Number of Magnitude of violations (`)

Days to Liquidity
maturity violations violations
(Percentage) Mean S.D. Q1 Q2 Q3 (Percentage) Mean S.D. Q1 Q2 Q3
Thinly 43 6
1739 2496 262 933 2091 2289 3168 643 1242 1944
traded (100) (100)
Moderately
Zero NA NA NA NA NA Zero NA NA NA NA NA
61–90 traded
Days Highly
Zero NA NA NA NA NA Zero NA NA NA NA NA
traded
43 6
Overall 1739 2496 262 933 2091 2289 3168 643 1242 1944
(1.48) (0.85)
In the table, S.D., Q1, Q2 and Q3 denote standard deviation, first quartile, second quartile (i.e., median) and third quartile, respectively.
 Testing Lower Boundary Conditions for the S&P CNX Nifty Index Options ● 51

registered in this category). Equally revealing observation is that the majority

of violations belong to the relatively lower levels of liquidity. Empirically, for
the call options, one-fourth of such violations belonging to the highly liquid
contracts seem to be exploitable (though relatively lower magnitude of profits)
in the presence of bid-ask spread. In contrast, nearly 50% of such violations
in the case of put options, per-se, could be exploitable, as the second quartile
shows a profit of ` 343 and more. Likewise, one-fourth of the moderately
traded category for such options seem to be exploitable, as the third quartile
offers profit of more than ` 462 and 368 in the cases of call and put options,
respectively. In sum, nearly 3% and 2% of violations which had 8–3 days to
maturity can be designated as exploitable in light of the profit magnitude
they offer and the bid-ask spread costs, respectively, in the cases of call and
put options.
In addition to the above, the last two levels of time to maturity, i.e. 31–60
and 61–90 days to maturity clearly depicts lack/negligible number of the
highly and moderately traded contracts for put options. To put it in other
words, the violation in the case of put options having 31–90 days to maturity
belong to the thinly traded category and, therefore, can safely be referred to as
unexploitable. Similarly, in case of call options, the contracts that are having
61–90 days to maturity do not contain a single moderately/highly traded
options contract and, thus, are unexploitable; however, as far as violations
with maturity period of 31–60 days are concerned, negligible number of
violations with moderately and highly traded category can be identified. Such
violations, per se, seem to be exploitable. However, the relatively long holding
period and relatively less trading in such options could have been the reasons
for their unexploitability.
In short, as far as the exploitability of the observed mispricing signals
is concerned, the results regarding maturity are in line with those in case
of liquidity, as the majority of violations pertain to the relatively illiquid
categories for all the four levels of maturity, for call as well as put options.

3.5.2 Statistical Significance of the Differences in the Magnitude of

Violations
From the above findings, it can be observed that there seems to be a difference
in the mean magnitudes of violations among the specified levels of liquidity for
call as well as put options. To validate the finding statistically, a well known
statistical test—Analysis of Variance (ANOVA)—was proposed initially.
However, before applying the test statistics on the data, the main assumption
of ANOVA, i.e. the samples have been drawn from a normally distributed
population has been tested using the one-sample Kolmogorov–Smirnov
statistics. The results are summarized in Table 3.4.
52 ● Derivative Markets in India

Table 3.4 Summary of the One-sample Kolmogorov–Smirnov Statistics to Assess the Normality
Options Call options Put options
Number of observations 2892 707
Mean 1057.85 1351.98
Normal parameters (a), (b)
Std. deviation 2426.71 2965.31
Absolute 0.33 0.32
Most extreme differences Positive 0.30 0.28
Negative −0.33 −0.32
Kolmogorov–Smirnov Z 17.83 8.62
Asymp. Sig. (2-tailed) 0.000 0.000
(a) Test distribution is normal; (b) Calculated from data.

Since the results depict severe departure from the normality (revealed by
the Kolmogorov–Smirnov, KS, statistics), ANOVA cannot be applied, as it
requires data to follow the normal distribution. Therefore, the differences
have been analysed using a non-parametric statistics that does not require
the data to follow any specified distribution. The test statistics applied in the
present study is Kruskal–Wallis (H-statistics) test, which is a non-parametric
substitute for the one-way ANOVA. In addition, Dunn’s multiple comparison
test has been used for post hoc analysis of all possible pairs in the analysis. The
results of H-statistics and Dunn’s test for the differences across the specified
levels of liquidity are summarized in Tables 3.5 (a) and (b), respectively.
Table 3.5(a) Kruskal–Wallis (H-statistics) Test for the Differences in Means (Magnitude) of
Violations across the Specified Levels of Liquidity for Call and Put Options,
June 2001–07
Call options Put options
Test statistics
Liquidity Rank Rank Test statistics (a), (b)
(a), (b)
Mean Chi- Mean Chi-
N df Sig. N df Sig.
rank square rank square
Thinly
2523 1502.72 654 359.20
traded
Moderately
278 1096.80 91.568 2 0.000 46 288.07 5.688 2 0.058
traded
Highly
91 956.24 7 301.14
traded
(a) Kruskal–Wallis Test (b) Grouping variable—liquidity.
The significance value given in Table 3.5 (a) clearly indicate that the
difference among the specified levels of liquidity is significant at even 1% level
of significance for call options; however, the difference could not be supported
statistically in the case of put options at 5% level of significance.
As a result of significant results of H-statistics in case of call options, the post
hoc diagnosis, i.e. Dunn’s multiple comparison test, has been carried out to find
the pair-wise differences. Notably, the post hoc analysis has not been applied in
 Testing Lower Boundary Conditions for the S&P CNX Nifty Index Options ● 53

case of put options as the differences amongst the specified levels of liquidity
were not found to be significant at the first stage itself, i.e. the Kruskal–Wallis
test. The results of the post hoc analysis in the case of call options summarized
in Table 3.5 (b) signify that average magnitude of violations for the thinly
traded options is significantly different from that for the moderately traded,
as well as highly traded options. However, average magnitude of moderately
traded options is not significantly different from that of highly traded options
at 5% level of significance.
Table 3.5(b) Dunn’s Test for the Multiple Comparisons amongst the Specified Levels of
Liquidity for the Call Options, June 2001–07
Call Options
Dunn’s multiple
Difference in Significant Summary
comparison test
rank sum (P < 0.05)
Thinly traded
vs. 405.9 – Yes
Moderately traded
Thinly traded
vs. 546.5 – Yes
Highly traded
Moderately traded vs.
140.6 Not significant No
Highly traded

In operational terms, the results imply that the average magnitude of

violations for exploitable options contracts is significantly different from those
for options contracts that can be designated as unexploitable. Empirically, the
finding validates that the magnitude of exploitable violations is significantly
less than that for unexploitable options. It demonstrates a good sign for
the market as the truly exploitable mispricing opportunities were meager
in magnitude, and significantly noticeable violations existed only in the
cases of contracts that could not be exploited due to lack of liquidity in such
options.

3.5.3 The Learning Curve

This part of the chapter attempts to test the learning hypothesis for call as well
as put options markets in Indian context over the period under reference. The
hypothesis warrants improvement in the market efficiency over the years as
investors are expected to be more familiar/experienced with the new market
year after year and, thus, are expected to behave more rationally in pricing
the options contracts. To put it explicitly, it has been hypothesized that the
mispricing signals should depict a declining trend for call as well as put
options. The examination of the violations over the years under reference is
similar with the study carried out by Mittnik and Rieken (2000a) in the context
of German index options market.
54 ● Derivative Markets in India

In this respect, the results (summarized in Table 3.6 and Fig. 3.1) indicate
that the percentage of violations in call options has shown a decreasing trend in
the first year (from 12% to nearly 8%). However, it remained almost persistent
for next four years as the percentage of violations kept on hovering around 8%
and showed a declining trend in the last year, as the percentage of violations
reduced to nearly 4% of the total observations analysed for the year. That is,
the violations pertaining to the call options have shown an improvement in the
first year of trading; however, no significant improvement could be recorded
during the next four years of the trading. Notwithstanding the persistence in
the percentage of violations, the last year under reference, i.e. 2006–07, has
shown considerable improvement in the state of market efficiency.
Table 3.6 Year-wise Frequencies of the Violations of LBC for Call and
Put Options, June 2001–07
Call options Put options
Year No. of No. of No. of No. of
Percentage Percentage
observations violations observations violations
(%) (%)
analysed observed analysed observed
2001-02 3496 422 12.07 2557 114 4.46
2002-03 3898 301 7.72 3376 151 4.47
2003-04 7173 455 6.34 6189 228 3.68
2004-05 7010 629 8.97 6702 61 0.91
2005-06 9323 735 7.88 7931 92 1.16
2006-07 9398 350 3.72 8416 61 0.72
Total 40298 2892 7.18 35171 707 2.01

14.00%
12.00%
10.00%
8.00%
6.00%
4.00%
2.00%
0.00%
2001–02 2002–03 2003–04 2004–05 2005–06 2006–07

Figure 3.1 Percentage of violations in relation to the total observations

analysed across the years for the call options.

In contrast, the percentage of violations over the years has shown a clearly
declining trend in the case of put options except for the first two years
of trading, where the percentage of violations remained nearly at the same
level (as is evident in Table 3.6 and Fig. 3.2). The third year of trading
showed a modest learning effect amongst the investors, as the percentage of
violations registered a decline of less than 1%. However, it may be noted that
 Testing Lower Boundary Conditions for the S&P CNX Nifty Index Options ● 55

the next year of trading registered an exemplary decline, as the percentage

of violations reduced to just nearly 1%. Moreover, the next two years have
shown favorable response and the percentage of violations finally reached
the level of 0.72%.

5.00%
4.50%
4.00%
3.50%
3.00%
2.50%
2.00%
1.50%
1.00%
0.50%
0.00%
2001–02 2002–03 2003–04 2004–05 2005–06 2006–07

Figure 3.2 Percentage of violations in relation to the total observations analysed

across the years for the put options.

Based on these observations, it may be appropriate to conclude that the

put options market has clearly demonstrated the learning effect amongst the
market participant as there has been a considerable decrease in the percentage
of violations, which finally reduced to 0.72% of observations analysed in the
last year. This clearly supports the learning hypothesis in the put options
market. In contrast, the call options market, however, has shown a decline
but the persistence of violations across four years of the study and a relatively
significant percentage of violations in the last year (i.e. 3.72% in 2006–07) do
not provide very convincing evidences on the learning behavior in the market.
The so observed differences in the trends of violations in the two markets have
been explained in the next subsection.
Also, the magnitude of the violations across the years has been examined
in order to have a better idea about the exploitability of violations. The results
in this regard have been summarized in Table 3.7. It is satisfying to note that
the first three (2001–02 to 2003–04) years of trading in the options market had
zero/negligible number of violations belonging to the highly traded category
for both call and put options.
Moreover, the violations traded in moderately traded category remained
negligible in case of put options for the first two years (2001–02 to 2002–03) of
trading. In contrast, though such violations for call options have shown higher
frequencies compared with those in the case of put options, the magnitude
of profits they offered can be referred to as unexploitable in the presence
of bid-ask spread. In the year 2003–04, the number of violations under the
moderately traded category has been nearly equal for the call and put options.
As far as the exploitability of such violations is concerned, nearly one-fourth
 Table 3.7 Year-wise Magnitude of Violations of the LBC and their Sub-classification as per the Specified 56
Levels of Liquidity in Indian Options Market for Call and Put Options, June 2001–07
●

Call options Put options

Years Liquidity Number of Magnitude of violations (INR) Number of Magnitude of violations (INR)
violations Mean S.D. Q1 Q2 Q3 violatious Mean S.D. Q1 Q2 Q3
387 111
Thinly traded 402 480 115 288 516 434 696 71 154 395
(91.71) (97.37)
34 3
Moderately traded 149 117 44 137 226 63 16 54 56 69
(8.06) (2.63)
Derivative Markets in India

2001–02
1
Highly traded 67 NA NA NA NA Zero NA NA NA NA NA
(0.24)
422 114
Overall 381 466 108 264 497 424 690 62 144 394
[14.5.0] [16.12]
264 143
Thinly traded 337 469 81 213 442 351 475 62 212 396
(87.71) (94.70)
32 7
Moderately traded 145 125 61 78 224 131 115 32 105 244
(10.63) (4.64)
2002–03
5 1
Highly traded 225 123 163 243 335 87 NA NA NA NA
(1.66) (0.66)
301 151
Overall 314 445 73 1.92 419 33.9 466 61 211 372
[10.41] [21.36]
422 201
Thinly traded 928 1978 130 333 822 1410 2053 218 562 1467
(92.75) (88.16)
27 24
Moderately traded 188 252 52 102 191 466 823 103 278 452
(5.93) (10.53)
2003–04
6 3
Highly traded 60 42 35 40 104 334 252 192 249 433
(1.32) (1.32)
455 228
Overall 874 1916 121 309 768 1294 1970 1970 514 1273
[15.73] [32.25]
 Call options Put options
Years Liquidity Number of Magnitude of violations (INR) Number of Magnitude of violations (INR)
violations Mean S.D. Q1 Q2 Q3 violatious Mean S.D. Q1 Q2 Q3
505 60
Thinly traded 974 1515 149 374 1054 1013 1294 162 441 1302
(80.29) (98.36)
76
Moderately traded 381 263 161 378 542 Zero NA NA NA NA NA
(12.08)
2004–05
48 1
Highly traded 266 327 74 140 328 598 NA NA NA NA
(7.63) (1.64)
629 61
Overall 848 1387 137 344 839 1006 1284 166 443 1301
[ 21.75] [8.63]
637 86
Thinly traded 1436 2551 230 565 1324 3827 6314 535 1327 3541
(86.67) (93.48)
82 5
Moderately traded 473 599 105 323 629 243 325 85 87 217
(11.16) (5.43)
2005–06
16 1
Highly traded 259 242 95 148 398 423 NA NA NA NA
(2.18) (1.09)
735 92
Overall 1303 2408 202 532 1190 3595 6166 402 1271 3457
[25.41] 113.01]
308 53
Thinly traded 2912 5173 421 925 2761 2404 3102 445 1373 2769
(88,00) (86.89)
27 7
Moderately traded 449 431 180 317 545 5847 6535 1139 2171 10089
(7.71) (11.48)
2006–07
15 1
Highly traded 404 280 227 337 612 281 NA NA NA NA
(4.29) (1.64)
350 61
Overall 2615 4920 323 770 2221 2764 3733 490 1404 3047
[12.10] 18.631
Testing Lower Boundary Conditions for the S&P CNX Nifty Index Options

1. Figures in parentheses indicate percentage.

●

2. In the table, S.D., Q1, Q2 and Q3 denote standard deviation, first quartile, second quartile (i.e., median) and third quartile, respectively.
57
58 ● Derivative Markets in India

of the violations seem to be exploitable in the presence of bid-ask spread for

put options, as suggested by the third quartile; however, such options are not
exploitable in the case of call options.
In the next three years of the study (i.e. 2003–04 to 2006–07), the frequency
of violations in the case of put options belonging to highly and moderately
traded category has been negligible, and majority of violations fall into the
thinly traded category, which can be designated as unexploitable. The finding
further corroborates the earlier finding that the put options market has shown
considerable improvement and therefore clearly demonstrated the learning
process amongst the market participants. In contrast, the number of violations
for highly and moderately traded call options has been significant for the
years 2003–04 to 2006–07. Moreover, nearly half of the violations belonging to
moderately traded category can be labelled as exploitable for all the three years
as the second quartiles suggest a profit of more than ` 317 per lot for rest of the
50% of observations. Similarly, nearly one-fourth of the violations pertaining
to highly traded category seem to be exploitable for the fourth and fifth year;
nearly 50% of such violations appear to be exploitable in the last year.
In sum, the call options market has shown some initial evidences of lack of
warranted improvement in the market efficiency unlike put options, which
have shown considerable improvement.

3.5.4 Comparison of Call and Put Options

In order to ascertain the levels of pricing efficiency in the two markets,
namely call options market and put options market, a comparison has been
drawn between these markets. The number of violations is 2,892 out of total
observations of 40,298 in the case of call options. Likewise, 707 violations have
been observed out of 35,171 observations in put options. As far as the frequency
of violations is concerned, the call options market seems to be more inefficient
compared with put options market as the number as well as the percentage
(7.18% of total observations analysed in call options compared with 2.01%
in put options) of violations are higher vis-à-vis put options. Moreover, the
percentage of violations over the period under reference has demonstrated a
persistent pattern in case of call options market unlike the put options market,
which has shown a considerable decline in the percentage of violations over
the years under reference.
The results based on the absolute figures are more revealing as all the
exploitable violations seem to be negligible for put options; however, such
violations in case of call options remained significant. It is eloquently borne
out by the fact that there are only 13 cases for put options whereas more than
90 for call options (one-fourth of the total violation in moderately and highly
liquid category for call as well as put options as suggested by third quartiles,
 Testing Lower Boundary Conditions for the S&P CNX Nifty Index Options ● 59

Table 3.2) that can be designated as exploitable; all the remaining violations
might have remained unexploited due to the lack of liquidity. Our finding
that the LBC for call options is violated more frequently compared with put
options is similar with one documented by Puttonen (1993), a study done in
the Finnish index options market.
In this respect, explaining the higher frequency of violations in case of
call options, Mittnik and Rieken (2000a) opined that selling the asset short,
particularly when the asset is an index, becomes very difficult. Notably, the
strategy needed to exploit the arbitrage on account of the violations of LBC
in case of call options requires shorting of the underlying asset unlike that in
case of put options, which requires a long position in the underlying asset. In
view of this, the comparatively higher frequency of violations in the case of
call options can be reasonably attributed to the short-selling constraint in the
Indian securities market for the period under reference.

3.5.5 Violations Using Futures Market

Besides, an attempt has been made to identify the violations of LBCs using
futures prices of the underlying assets. This has been done in view of the
facts that
(i) the facility of short-selling the financial assets was absent in the Indian
securities market during the period under reference and futures market
easily provides this facility;
(ii) in case of index options market, it becomes very difficult to execute the
operations required to be taken in the underlying’s cash market (i.e.
short-sell/ purchase of the index) as all the shares in the index need to
be sold/ purchased in the proportion in which these have been included
in the index, and
(iii) transactions costs associated with trading in futures market are relatively
less compared with the spot market.
The results on the violations of LBCs using futures contracts are reported
in Table 3.8.
The results summarized in Table 3.8 reveal that the number of violations of
LBC remained quite high for call options compared with those for put options
even when the futures market has been used. Notably, unlike identifying
violations of LBCs using spot market, short-selling constraint could not be a
correct explanation to the higher frequency of violations in call options since
to exploit the arbitrage opportunities using futures market, the arbitrageur
does not have to short the stock basket; rather, he needs to sell the futures
that is easily possible.
60 ● Derivative Markets in India

Table 3.8 Violations of the LBC Using Spot Values and Futures
Prices, June 2001–07: A Comparison
Using spot values Using futures prices
Particulars Call Put Call Put
options options options options
Total number of observation analysed 40,298 35,171 40,298 35,171
Total number of violations observed 7019 1544 3593 1815
before transaction costs (17.42) (4.39) (8.49) (5.16)
Total number of violations observed 2892 707 2838 1492
after transaction costs (7.18) (2.01) (7.04) (4.24)
Figures in parentheses denote percentages.

Moreover, it may be noted that use of futures market has shown a negligible
improvement in the call options market as the number of violations has slightly
decreased from 2892 to 2838; however, the put options market has registered
a counterintuitive increase in the number of violation from 707 to 1492. The
reason behind the decrease (increase) in the number of violation for the call
(put) options could be the underpricing of futures contracts. Moreover, another
reason for the increase in the number of violations in the case of put options
could be the comparatively lower transaction costs in F&O segment. However,
there should have been an increase in the case of call options as well to assign
lower transaction costs as a major reason. Notably, there has been a negligible
decline in the case of call options, and therefore, the lower transaction costs
cannot be assigned as the sole reason behind this development.
The picture becomes clearer when we identify the violations without
considering any transaction cost for both the strategies, i.e. using the spot
market and futures market for identifying the violations of LBCs (as reported
in Table 3.8). The results demonstrate that the use of futures market has caused
a decrease (increase) in the frequency of violations pertaining to call (put)
options compared with those using spot market. In view of this, it would be
reasonable to conclude that the above-mentioned development in the state
of options market efficiency can be designated to the underpricing of futures
market. This may further be traced to the fact that as far as the correction
of underpricing of the futures market is concerned, the short-selling of the
stock basket is needed to exploit such arbitrage opportunities and to restore
equilibrium in the futures market. In short, the underpricing of the futures
can, therefore, be attributed to the fact that short-selling have been banned
during the period under reference in Indian securities market.
In sum, it may be concluded that the futures market has failed to provide an
equally good substitute for shorting the assets in the absence of short-selling
facility in the Indian securities market during the period under reference.
Thus, the futures market could not succeed in restoring equilibrium in the
options market.
 Testing Lower Boundary Conditions for the S&P CNX Nifty Index Options ● 61

In view of this, it is reasonable to conclude that the indirect impact of the

short-selling constraints on the efficiency of the options market on account of
the interrelationship of the index options and index futures market has been
one of the major reasons, amongst others (e.g. liquidity), for the existence of
mispricing signals in the Indian options market. In short, the impact of short-
selling constraints cannot be ignored even if the violations are identified using
futures contracts, as the efficiency of futures market does impact the efficiency of
options market and, which in turn, can be ensured when short-selling is allowed.

3.6 CONCLUDING OBSERVATIONS

The study attempts to test the LBC for the S&P CNX Nifty index options prices
using the futures prices on the same index in the Indian securities market. The
results of the study are, more or less, in line with those drawn in the case of
US market (e.g. concentration of violations in 0–7 days to maturity category),
except for magnitude and frequency of violations, which have been observed to
be more pronounced in Indian options market, alike the Finnish index options
market. Also, the frequency of violations in call options have been found to be
more pronounced compared with that in put options. The violation of lower
boundary indicates underpricing of options in Indian securities market. The
finding that the options are underpriced is consistent with that of Varma
(2002), a study carried out in Indian context.
Though the frequency of violations remained quite high in Indian options
markets compared with that in the US market, the exploitability of such
violations remained confined to a negligible number on account of the dearth
of liquidity. In other words, a significant number of violations remained
unexploitable, plausibly on account of the lack of liquidity and the direct as well
indirect (through the futures market) impact of short-selling constraints.
The study is equally revealing as far as the behaviour of the investors
dealing with the options market in India is concerned. It has been observed
that the number of violations in call options market has been persistent instead
of showing a warranted declining trend. In other words, it implies that the
irrationalities in the behaviour of investors, particularly in call options market,
could not improve significantly over the years. However, it is satisfying to
note that the put options market has behaved in a way that is consistent with
the learning hypothesis, i.e. the number of violations has reduced with the
passage of time. Thus, the findings indicate that the put options market is
emerging to be more efficient vis-à-vis the call options market.
Another notable finding of the study is that the futures market could not
provide an equally good substitute for the short-selling facility in the Indian
securities market. This is evident from the fact that the number of violations
in case of call options remained nearly same even when the futures market
has been used to identify the violations. On the other hand, the put options
62 ● Derivative Markets in India

market has shown an increase in the number of violations for such an arbitrage
strategy. Such a development can be traced to the fact that the futures
themselves remained underpriced in the absence of short-selling facility in the
market and, therefore, registered a negligible decrease in the case of call options
and an increase for the put options market. In sum, it may be reasonable to
infer that the futures market failed to restore equilibrium in options market
in the absence of short-selling facility in Indian securities market for the time
period under reference.
Therefore, in short, it is reasonable to conclude that majority of violations
in call options could not be exploited on account of the existing market-
microstructure in India during the period under reference (i.e. short-selling
constraint that caused underpricing in futures to persist). Besides, the dearth
of liquidity in the options market appears to be another major constraint to
arbitrageurs, in case of call as well as put options market, in view of the fact
that a vast majority of violations occurred in the thinly traded category.

END NOTES
1. Nonsynchonous trading refers to the phenomenon of different timings
of closing transactions in the two markets (the options market and the
underlying’s cash market in this case).
2. In the study, the liquid options quotations have been defined as those
quotations that have at least one contract traded. Though the definition
of liquid contracts is not useful in ensuring exploitability of arbitrage
opportunities on account of the high bid-ask spread for such options, this
has been done to gauge the total number of violations in Indian options
market. Moreover, while ensuring the exploitability of the arbitrage
opportunities, due consideration has been given to the liquidity.
4 C H A P T E R

A Test of Put–Call Parity

Relationship on S&P CNX
Nifty Index Options Market

4.1 INTRODUCTION
Efficiency of the options markets is of great importance to the academicians
as well as practitioners. Well-functioning financial markets are vital to a
thriving economy, as these markets facilitate price discovery, risk hedging and
allocation of capital to its most productive usage. Inefficiency of a financial
market (e.g. options market) indicates that it is not performing the best possible
job at above-mentioned important functions (Ackert and Tian, 2000). Therefore,
it is imperative to test the state of options market efficiency, especially when
the market is in its nascent stage of development.
In view of this, this chapter attempts to investigate the market efficiency
of the Indian index options market by testing the Put–Call Parity (PCP)
relationship on daily closing observations of call and put options. The PCP
relationship in line with the lower Boundary conditions (LBCs) is a necessary
but not sufficient condition for the efficiency of the options market. However,
it is a more rigorous condition on options market efficiency compared with the
LBCs approach, as it requires the relationship between call and put options to
hold. The efficiency of the market, in this respect, connotes pricing efficiency.
The PCP for the option prices (as specified in the literature) has been used
extensively as a tool to gauge the efficiency of options market across the globe.
The violation of PCP implies that the call (put) options are not priced correctly
in relation to the corresponding put (call) options and, therefore, indicates
the arbitrage opportunities in the market that are expected to be absent in an
efficient market.
Moreover, the futures market on the same underlying asset (S&P CNX Nifty
Index) has also been used to test the PCP relationship, i.e. Put-Call-Futures
Parity (PCFP). This has been done in view of the fact that (i) the futures
market is a better alternative in exploiting the arbitrage opportunities when
underlying asset is an index; (ii) it is generally argued that the futures market
helps in doing away with the short-selling constraint as a futures can easily be
64 ● Derivative Markets in India

shorted; and (iii) it costs an investor less to exploit the arbitrage opportunities
through futures market because of the lower transaction costs associated with
it and the leverage it provides. In sum, the futures market becomes a more
convenient and profitable route in the case of index options to exploit the
arbitrage opportunities.
Notably, the use of futures prices on the same underlying asset instead of
spot prices essentially makes this approach a test of joint market efficiency,
as opined by Fung et al. (1997). At the same time, use of the futures prices
facilitates in assessing the degree of pricing interrelationships between the
different derivative instruments/markets traded in the financial market
(Lee and Nayar, 1993). In other words, this approach helps in addressing
the question whether market participants consider important pricing
interrelationships while pricing the index options. The scope of the present
study has been confined to the pricing interrelationship between index options
and index futures.
In sum, the very purpose of using futures market has been to examine its
role in the absence of short-selling facility in underlying’s market. In other
words, whether the futures market could work as an equally good alternative
to short-selling facility and, therefore, help in restoring equilibrium in the
options market even in the absence of short-selling facility.
The use of futures market imposes one restriction on the otherwise model-
free approach as it assumes ‘cost-of-carry’ model to hold. Therefore, it would
not be appropriate to call this approach as a ‘model-free’ approach unlike the
test procedure which uses spot values. However, the approach still remains
less restrictive compared with those based on certain pricing models, e.g.
Black and Scholes (1973), which assumes that the stock price and volatility
are governed by some stochastic processes.
The method applied to test the efficiency of the options market is in line
with other studies conducted in different markets for the same purpose. The
test procedure has taken care of the dividends expected from the underlying
asset during the life of the options. Moreover, the transaction costs (excluding
bid-ask spread) have been incorporated while identifying the violations of PCP
relationship. Though the analysis has been conducted ignoring the bid-ask
spread, the results have been interpreted cautiously. This has been considered
in view of the fact that the bid-ask spread plays a very important role in
assessing the options market efficiency as it results in significant transactions
costs (Baesel et al.,1983; Phillips and Smith, 1980).
The violations or mispricing signals observed from the test procedures have
further been classified as per the specified levels of ‘liquidity’ and ‘time to
maturity’ in order to draw some meaningful results from such violations. The
classification so made facilitates a logical explanation to the exploitability of
such violation, which is very crucial in assessing the efficiency of the market.
 A Test of Put–Call Parity Relationship on S&P CNX Nifty Index Options Market ● 65

This has been done in view of the fact that mere presence of violations does
not indicate market inefficiency; it is the exploitability and persistence of such
violations that pose serious threat to the market efficiency.
Besides, the learning behaviour of the investors in options markets has also
been examined. This has been done by analysing the number of violations
vis-à-vis the number of observations analysed across the subsequent years,
for both the call and put options. The learning hypothesis warrants that the
number of violations should go down over the years. This has been proposed
with intent to gauge the desired improvements in the understanding of the
options market amongst market participants.
The rest of the chapter is divided into four main sections. The Section 4.2
of this chapter discusses PCP relationship using spot as well as futures prices
on the same index. The data has been summarized in Section 4.3. Section 4.4
presents analysis and empirical results. The chapter ends with the concluding
observations in Section 4.5.

4.2 PUT–CALL PARITY (PCP) RELATIONSHIP

The PCP theory that constitutes a central role in options pricing, was first
put forth by Stoll (1969) and Merton (1973b). Stoll (1969) tested this theory
on over-the-counter (OTC) market in the American context for the very first
time. This theory using ‘conversion mechanism’ establishes a relationship
between the premium on a call option and that of the corresponding put
option having same characteristics (in terms of strike price, trade date and
maturity date). The conversion mechanism allows converting a call (put) in
to a put (call) using long and short positions in call, put and their underlying
asset. The process of conversion is risk-less as it results into a completely
hedged position. Moreover, the mechanism is assumed costless in the absence
of transaction costs. In literature, it has been documented that this theory
essentially holds for the European options and not for the American options
in its strict sense.
Notably, the PCP relationship indicates the equilibrium price of a call
(put) option given the price of a corresponding put (call) option. That is, it
indicates whether the call options are priced correctly in relation to the price
of a corresponding put options and vice-versa. However, it does not comment
upon the correctness of their prices when seen individually. That is, the parity
relationship may hold even when both the put and call options are underpriced
or correctly priced or overpriced. Therefore, from the pricing viewpoint, it is
a necessary condition but not a sufficient one.
Given the importance of parity relationship in assessing the pricing efficiency
of the options market, a number of studies have been attempted to test the
options market efficiency (using this relationship) in various markets around
66 ● Derivative Markets in India

the globe. The model-free approach of this efficiency test, i.e. no dependence
on any particular options pricing model (which makes certain assumptions
about the movement of the underlying asset and volatility during the life of
the options, e.g. Black–Scholes model), makes it of particular importance to
assess the efficiency of options markets. Some of the studies that have used
this test to assess the efficiency of different options markets are Stoll (1969),
Merton (1973), Gould and Galai (1974), Klemkosky and Resnick (1979), Goh
and Allen (1984), Evine and Rudd (1985), Bodurtha and Courtdon (1986), Gray
(1989), Taylor (1990), Frankfurter and Leung (1991), Wilson and Fung (1991),
Finucane (1991), Brown and Easton (1992), Nisbet (1992), Marchand et al. (1994),
Sternberg (1994), Easton (1994), Kamara and Miller Jr. (1995), Wagner et al.
(1996), Berg et al. (1996), Broughton et al. (1998), Mittnik and Rieken (2000b),
Cavallo and Mammola (2000) , Li (2006), Zhang and Lai (2006).
Likewise, a few studies have used put-call-futures relationship (which is
the test of PCP relationship using futures prices on the same underlying asset
instead of spot market values) to examine the options market efficiency and
the degree of pricing interrelationship between the two markets. The leading
study in this respect, Lee and Nayar (1993), for the very first time proposed
this relationship and tested it in the context SPX options and S&P 500 futures
traded on Chicago Board Options Exchange (CBOE) and Chicago Mercantile
Exchange (CME) of USA. Other studies that used this framework to assess
options market efficiency include Fung and Fung (1997), Fung and Mok (2001),
Draper and Fung (2002), Vipul (2008), etc.

4.2.1 The PCP Relationship Using Spot Values

The PCP relationship using the spot values for the European type options
contracts is given in Eq. (4.1). The relationship given in the equation has been
established assuming no bid-ask spread.
Ct = [Pt + (It – e– r (T – t) K) ± TTCt] (4.1)
Where,
Ct is the market price of a call option at time t,
Pt is the market price of a put option at time t,
It is the level of underlying index (i.e. S&P CNX Nifty) at time t,
K is the strike price of the options contracts,
T is the expiration time of the option at the time when it was floated,
r is the continuously compounded annual risk-free rate of return,
TTCt is the Total transaction costs (i.e. transaction costs relating to trading
in options and spot market) at time t and
(T − t) is the time to maturity of the option at time t in years.
In addition to the assumption of no transaction costs, the Eq. (4.1) for parity
relationship also assumes that the underlying asset is payout protected, i.e. it
 A Test of Put–Call Parity Relationship on S&P CNX Nifty Index Options Market ● 67

is not expected to pay any dividends during the life of the option. However,
the assumption of payout protection is violated quite frequently as almost
all the financial assets pay dividends. Equation (4.1) needs to be modified by
incorporating the effect of dividends. Empirically, the treatment of dividends
in the test varies based on the assumption made about the payment of
dividends. In this respect, Klemkosky and Resnick (1979), for the very first
time, incorporated effect of absolute amount of dividends to modify the parity
relationship.
A number of studies in literature have used discrete amount of dividends
to gauge the effect of dividends on the parity relationship. However, in the
present chapter, since the underlying asset is a stock index (S&P CNX Nifty
index), which is based on 50 stocks, the dividend yield has been used to
incorporate the effect of dividends on the parity relationship following Fung
and Fung (1997) and Li (2006). This has been done in view of the fact that
incorporation of discrete dividends becomes difficult when the underlying
asset is a stock index. Also, a stock index can be treated as a financial asset
that generates a continuous stream of dividends (Guo and Su, 2006). The PCP
relationship for the S&P CNX Nifty index options, assuming that the index
is paying continuously compounded annual dividend yield (d), is given in
the Eq. (4.2).
Ct = [Pt + (– e– d (T – t) It – e– r (T – t) K) ± TTCt] (4.2)
The testable form of the Eq. (4.2), which constitutes the basis of the present
study in assessing the efficiency of the options market in India, is given in the
next sub-section.

4.2.1.1 Testable form of the PCP relationship using spot values

Equation (4.2) given above has been rearranged in order to make it testable to
gauge the efficiency of the options market. The testable forms are given in the
Eqs (4.3) and (4.4) for identifying overpricing and underpricing of put options
relative to the corresponding call options with same contract specifications
(assuming that the call is correctly priced).
e tOverpriced = PtMarket – CtMarket – ( e – d (T – t ) * It – e – r (T – t ) * K ) + TTCt (4.3)

e tUnderpriced = CtMarket – PtMarket + ( e – d (T – t ) * It – e – r (T – t ) * K ) – TTCt (4.4)

In the above equations, e tOverpriced and e tUnderpriced are the absolute amount
of abnormal profits (ex-post) or mispricing signals, if the violation of PCP
takes place. A violation of PCP is recorded if e tOverpriced > 0 or e tUnderpriced > 0.
Though the presence of such profits is merely indicative of market inefficiency,
it should not be treated as a conclusive remark on the efficiency of the market,
as their exploitability needs to be examined to draw correct inferences about
the market efficiency.
68 ● Derivative Markets in India

Example 4.1 Exploiting arbitrage opportunity indicated by violation of PCP

condition: A case of relative underpricing of put option
On December 1, 2010, a call option contract on S&P CNX Nifty index with
strike price of ` 6200 and scheduled to expire on December 31, 2010, is currently
available at ` 200. At the same time, a put option contract with the same
characteristics is traded at ` 50. In the spot market, the index is currently being
traded at ` 6290. Suppose that continuously compounded risk-free rate of return
is 7.5% p.a., and each transaction in F&O segment is subject to transaction cost
of 0.05% on notional value of the contract, i.e. (strike price + premium) ¥ size
of the contract. An option contract on Nifty includes 50 Nifty.
Solution As mentioned in the discussion on PCP relationship, corresponding
price of a put option contract (given the price of a call option with the same
characteristics) at any given point in time should be
PtTheoretical = CtMarket – ( It – e r (T t ) K ) + TTCt
– –

In the above-mentioned equation, it is important to note that the transaction

cost has been subtracted. This has been done in view of the fact that we are trying
to trace underpricing of the put option, and therefore, we need to arrive at the
lower bound of the price range that will exist in the presence of transaction
costs.
Based on the above-mentioned relationship, the theoretical price of a put
option, given that a corresponding call option with the same characteristics is
currently traded at ` 200, should be
= {200 – (6290 – e– (0.075 ¥ 31/365) ¥ 6200)
– (2 ¥ 6200 + 200 + 50) ¥ 0.0005} = ` 64.311
Since current market price of the put option (` 50) is lower than that
warranted by PCP relationship; it indicates an arbitrage opportunity. In case
such an opportunity appears, following steps can be taken to ensure arbitrage
gains.
Steps required now (on spot):
All steps required now need to be taken simultaneously to lock-in arbitrage
profit. This will hold true for Examples 4.1–4.4.
Step 1: Purchase the put option as it is undervalued; and sell the call option as
it seems to be overvalued in relation to put option. This will lead to synthetic
short-position in the underlying asset.
Step 2: Take long position in the underlying asset.

1
It is important to note that the transaction costs related to spot market as well as dividend yield
have been ignored in determining the relative price of the put option in order to avoid further
complexity.
 A Test of Put–Call Parity Relationship on S&P CNX Nifty Index Options Market ● 69

Step 3: Borrow ` 6,146.33 (` 6290SPOT PRICE + ` 50PUT OPTION PRICE – ` 200CALL

OPTION PRICE + ` 6.325TRANSACTION COST) at risk-free rate for the remaining life
of the contract.
Steps required at maturity:
Step 4: Square off your long position in the underlying asset.
In this respect,(a) call option will be exercised against you, and you have
to sell the asset (you are long in) at ` 6200, in case underlying asset is being
traded at more than ` 6200. (b) On the contrary, if the underlying is traded at
a lower price, e.g. ` 6150; you can sell it at ` 6200 by exercising the put option.
In either case, you will end up receiving ` 6200 by selling the underlying asset
you are long in.
Step 5: Pay back the borrowed sum along with the accrued interest. You need
to pay a sum of ` 6146.33 ¥ e(0.075 ¥ 31/365); that is, ` 6185.61.
From the above, it is evident that, in any case, you will receive ` 6,200 by
selling the underlying asset. Since you need to pay ` 6,185.61 for the initially
borrowed sum of money, you end up generating a profit of ` 14.39 (` 6,200 −
` 6,185.61). Thus, on the whole contract (which includes 50 Nifty), you will
earn an arbitrage profit of ` 719.50.

Example 4.2 Exploiting arbitrage opportunity indicated by violation of PCP

condition: A case of relative overpricing of put option
On December 1, 2010, a call option contract on S&P CNX Nifty index with
strike price of ` 6200 and scheduled to expire on December 31, 2010, is currently
available at ` 200. At the same time, a put option contract with the same
characteristics is traded at ` 90. In the spot market, the index is currently being
traded at ` 6290. Suppose that continuously compounded risk-free rate of return
is 7.5% p.a., and each transaction in F&O segment is subject to transaction cost
of 0.05% on notional value of the contract, i.e. (strike price + premium) × size
of the contract. An option contract on Nifty includes 50 Nifty.
Solution As mentioned in the discussion on PCP relationship, corresponding
price of a put option contract (given that a call option with the same characteristics
is currently traded at ` 200) at any given point in time should be
PtTheoretical = CtMarket – ( It – e r (T t ) K ) + TTCt
– –

In the above-mentioned equation, it is important to note that transaction cost

has been added while determining price of the put options. This has been done
in view of the fact that we are trying to trace the overpricing of put option, and
therefore, we need to consider upper bound of the price range.
Therefore, corresponding price of a put option, given that a corresponding call
optionwith the same characteristics is currently traded at ` 200, should be
70 ● Derivative Markets in India

= {200 − (6290 − e−(0.075 ¥ 31/365) ¥ 6200)

+ (2 ¥ 6200 + 200 + 80) ¥ 0.0005} = ` 76.972
Since the current market price of the put option (` 90) is greater than that
required by PCP relationship; it indicates an arbitrage opportunity. In case such
an opportunity emerges, following steps can be taken to ensure arbitrage gains.
Steps required now (on spot):
Step 1: Sell the put option as it is overvalued and buy the corresponding call
option. This will lead to synthetic long position in the underlying asset.
Step 2: Sell the underlying asset short.
Step 3: Invest the amount received by constructing the above-mentioned
portfolio, i.e. ` 6,173.66 (` 6290STop price + ` 90Put option price – ` 200Call option price
– ` 6.34Transaction cost) at risk-free rate of return for the remaining life of the
contract.
Steps required at maturity:
Step 4: Liquidate your investment. You will realize a sum of ` 6173.66 ¥ e(0.075
¥ 31/365)
. That is, ` 6213.11.
Step 5: Square off your short position in the underlying asset.
For the purpose, (a) exercise the call option contract and buy the underlying
asset at ` 6200, in case the underlying asset is being traded at more than ` 6200.
(b) On the contrary, if the underlying is traded at a lower price, e.g. ` 6150; the
put option (that you sold initially) will be exercised against you. In a nutshell,
you will end up purchasing the underlying asset at ` 6200.
From the above, it is clear that, in any case, the amount required to square
off the short position is ` 6,200. Since you received ` 6,213.11 from your initial
investment, you end up generating a profit of ` 13.11 (` 6,213.11 − ` 6,200).
And, on the whole contract (which includes 50 Nifty), you will earn a profit
of ` 655.50.

4.2.2 The PCP Relationship Using Futures Prices

The test of PCP relationship using futures prices is in line with Lee and Nayar
(1993), Fung and Chang (1994), Fung et al. (1997), Fung and Fung (1997),
Fung and Mok (2001), etc. The PCP relationship for the options using the
corresponding futures prices (with the same maturity date) is given in the Eq.
(4.4). This condition is expected to hold in an efficient options market and,
therefore, has been used to test the options market efficiency.
Ct = [Pt + e– r(T – t) (Ft – K) ± TTCt*] (4.5)
In the above equations, Ft is the value of the S&P CNX Nifty futures (with
same expiration date as of the options under consideration) at time t and
2
It is important to note that the transaction costs related to spot market as well as dividend yield
have been ignored in determining the relative price of the put option in order to avoid further
complexity.
 A Test of Put–Call Parity Relationship on S&P CNX Nifty Index Options Market ● 71

TTCt is the Total transaction costs (i.e. transaction costs relating to trading in
options and futures contracts) at time t. Besides, all other variables are the
same as in Eq. (4.1).
With respect to the treatment of dividends, it may be noted that the futures
prices (in an efficient market) are expected to have impounded the effect of
dividends expected to be distributed during the life of the contract. And, it is
for this reason that the dividends expected from the underlying asset during
the life of the option have not been included as the underlying asset used in
the test is futures prices/values of the index instead of the spot values.

4.2.2.1 Testable form of the PCP relationship using futures prices

Equation (4.5) has been rearranged in order to make it testable to gauge the
efficiency of the options market. The testable form of the PCFP relationship
has been provided in the Eqs (4.6) and (4.7).
e tOverpriced = PtMarket – CtMarket – e r (T t ) * ( Ft – K ) + TTCt*
– –
(4.6)

e tUnderpriced = CtMarket – PtMarket – e r (T t ) * ( Ft – K ) – TTCt*

– –
(4.7)

In the above equations, e tOverpriced and e tUnderpriced are the absolute amount
of abnormal profits (ex-post) or mispricing signals, if the violation of PCP takes
place. A violation of PCP is recorded if e tOverpriced > 0 or e tUnderpriced > 0.
It may be noted that all equations relating to test of PCP using spot as well
as futures prices have been specified considering the transaction costs, but
assuming zero or negligible bid-ask spread. In view of this, there is always
a chance that the arbitrage opportunities suggested by these equations may
disappear in the presence of the bid-ask spread, especially for the options
traded relatively less frequently. Therefore, due consideration has been given
to the bid-ask spreads while interpreting the violations to draw inferences
regarding the market efficiency. The details on the transaction costs included in
the analysis have been summarized in the data section. On the contrary, given
the fact that the bid-ask spread for options is not included in the transaction
database provided by NSE and the difficulty to estimate such costs, it has
been excluded in the above equations. In operational terms, our study is in
line with that of Halpern and Turnbull (1985).

Example 4.3 Exploiting arbitrage opportunity indicated by violation of PCFP

condition: A case of relative underpricing of put option
In addition to the data given in Example 4.1, assume that a futures contract
on NSE CNX Nifty index, scheduled to expire on December 30, 2010, is traded
at ` 6330. Examine the arbitrage opportunity, if any. Determine the profit that
can be generated by exploiting this opportunity.
72 ● Derivative Markets in India

Solution As mentioned in the discussion on PCP using futures prices, the

corresponding price of a put option, given the price of a corresponding call option
with the same characteristics, should be
Pt = [Ct – e–r(T – t) (Ft – K) – TTCt*]
As we are aware from the previous examples that we need to subtract
transaction cost while tracing underpricing. Therefore, the relative price for
above-mentioned put option should be
= [200 – e(–0.075 ¥ 31/365){6330 – 6200}
– (2 ¥ 6330 + 6200 + 50) ¥ 0.0005] = ` 61.373
Since the price quoted in market (` 50) is lower than that suggested by parity
condition, it indicates an arbitrage opportunity. In case such an opportunity
appears in the market, following steps need to be taken to make arbitrage profit.
Steps required now (on Spot):
Step 1: Buy the put option at the current market price as it is undervalued and
sell the call option at the current market price. It will lead to a synthetic short
position in futures, which is traded at ` 6200 (strike price).
Step 3: Take long position in futures on the same index with the same maturity.
Step 4: From the above-mentioned portfolio, you will receive a sum of ` 140.54
(` 200Call option − ` 50Put option − ` 9.46Transaction cost). Invest it at risk-free rate for
the remaining time to maturity.
Steps required at maturity:
At maturity, price of the underlying asset may follow any of the following
three scenarios:
(i) less than the strike price of the option contracts,
(ii) more than the strike price but less than or equal to the current future
price (Ft) and
(iii) more than the current futures price (Ft).
Scenario 1: Less than the strike price of option contract, i.e. price of the
underlying index turns out to be less than ` 6,200. For example, at maturity,
the index is traded at ` 6150.
Step 4: You will exercise the put option as the current price of underlying asset
is less than strike price (` 6200) of the contract. However, the call option will
not be exercised against you. From your overall position in options market, you
will make a gain of ` 50 (` 6200 − ` 6150).
Step 5: On the contrary, you will incur a loss of ` 1804 (` 6330 - ` 6150) on
long position in futures, since you entered into a contract to buy the underlying
asset at ` 6330 which is currently traded at ` 6150 .
3
It may be noted that the margin needed to take a position in futures market and cost thereof
have been ignored in determining relative price of option contract in order to avoid further
complexity.
4
In settlement of futures contracts, it has been assumed that futures prices converge to the spot
price at maturity. This will hold true in case futures are priced correctly.
 A Test of Put–Call Parity Relationship on S&P CNX Nifty Index Options Market ● 73

In sum, you will incur a loss of ` 130 (` 180 − ` 50), in case price turns out
to be less than the strike price of options contract.
Scenario 2: More than the strike price of call option contract but less than
or equal to futures price. For example, at maturity, the index is traded at
` 6250.
Step 4: The call option will be exercised against you as the price is more than
strike price (` 6200) of the contract. As a result, you will incur a loss of ` 50. On
the other hand, you will not exercise put option given the market conditions.
Step 5: Similarly, you will incur a loss of ` 80 on the long position in futures.
In sum, you will incur a loss of ` 130 from you position in options and
futures markets.
Scenario 3: More than the price of futures contract (the price at which
we purchased the futures). For example, at maturity, the index is traded at
` 6400.
Step 4: The call option will be exercised against you as the price more than strike
price (` 6200) of the contract, and you will incur a loss of ` 200. On the other
hand, you will not exercise put option given the market conditions.
Step 5: However, you will gain ` 70 on the long position in futures.
In sum, you will incur a loss of ` 130.
Thus, from all the three possible scenarios for the price of underlying asset,
it is evident that you will end up with the loss of ` 130.
Step 6: Liquidate your investment, and you will have ` 141.44 (` 140.54 ¥ e
(0.075 ¥ 31/365)
).
From the above, it is clear that, in any case, you incur a loss of ` 130; at the
same time, you will receive ` 141.44 from your initial investment. In other words,
you make a profit of ` 11.44 (` 141.44 − ` 130). Thus, on the whole contract
(which includes 50 Nifty), you will end up with the profit of ` 572.

Example 4.4 Exploiting arbitrage opportunity indicated by violation of PCFP

condition: A case of relative overpricing of put option
In addition to the data given in Example 4.3, assume that the put option is
traded at ` 90 instead of ` 50.
Solution As mentioned in the discussion on PCP using futures prices, the
corresponding price of a put option, given the price of a corresponding call option
with the same characteristics, should be:
Pt = [Ct – e–r(T – t) (Ft – K) – TTCt*]
It is important to note that we have added transaction cost as we need the
upper bound of price range in order to trace overpricing.
Therefore, the relative price for above-mentioned call option should be
74 ● Derivative Markets in India

= [200 – e(–0.075 ¥ 31/365){6330 – 6200}

+ (2 ¥ 6330 + 6200 + 50) ¥ 0.0005] = ` 80.305
Since the price quoted in the market (` 90) is greater than that suggested
by the parity condition, it indicates an arbitrage opportunity. In case such an
opportunity appears in the market, following steps need to be taken to make
arbitrage profit.
Steps required now (on Spot):
Step 1: Sell the put option at the current market price as it is undervalued and
buy the call option at the current market. It will lead to a synthetic long position
in futures, which is traded at ` 6200 (strike price).
Step 2: Take short position in futures on the same index with the same
maturity.
Step 3: To create the above-mentioned portfolio, you need to borrow a sum of
` 100.52 (` 200Call option − ` 90Put option + ` 9.48Transaction cost) at risk-free rate for
the remaining time to maturity.
Steps required at maturity: At maturity, price of the underlying asset may
follow either of the following three scenarios:
(i) less than the strike price of the option contracts,
(ii) more than the strike price but less than or equal to the current future price (Ft) and
more than the current futures price (Ft).
Scenario 1: Less than the strike price of option contract, i.e. value of the
underlying index turns out to be less than ` 6,200. For example, at maturity,
the index is traded at ` 6150.
Step 4: The put option will be exercised against you as the price of the asset is
less than strike price (` 6200) of the contract. However, you will not exercise
the call option. From your positions in options market, you will incur a loss of
` 50 (` 6200 − ` 6150).
Step 5: On the other hand, you will make a profit of ` 1806 (` 6330 − ` 6150) on
short position in futures since you entered into a contract to sell the underlying
asset at ` 6330 at maturity. As the current market price in the spot market turns
out to be ` 6150, you will make a profit.
In sum, you will make a profit of ` 130 (` 180 − ` 50), in case price turns
out to be less than strike price of options contract.
Scenario 2: More than the strike price of call option contract but less than or
equal to futures price. For example, at maturity, the index is traded at ` 6250.
Step 4: You will exercise the call option as the price of the asset is more than
the strike price (` 6200) of the contract. Thus, you will make a profit of ` 50.
5
It may be noted that the margin needed to take a position in futures market and cost thereof
have been ignored in determining relative price of option contract in order to avoid further
complexity.
6
In settlement of futures contracts, it has been assumed that futures prices converge to the spot
price at maturity. This will hold true in case futures are priced correctly.
 A Test of Put–Call Parity Relationship on S&P CNX Nifty Index Options Market ● 75

On the other hand, the put option will not be exercised against you given the
market conditions.
Step 5: Similarly, you will make a profit of ` 80 on the short position in
futures.
In sum, you will make a profit of ` 130 from you position in options and
futures markets.
Scenario 3: More than the price of futures contract. For example, at maturity,
the index is traded at ` 6400.
Step 4: You will exercise the call option as the price of the asset is more than
the strike price (` 6200) of the contract. You will make a profit of ` 200. However,
the put option will not be exercised against you given market conditions.
Step 5: However, you will incur a loss of ` 70 on the short position in
futures.
In sum, you will gain ` 130.
Thus, from all the three possible scenarios for the price of underlying asset,
it is evident that you will make ` 130 or more.
Step 6: Pay back your borrowed sum (along with interest) of ` 101.16
(` 100.52 ¥ e (0.075 ¥ 31/365)).
From the above, it is clear that, in any case, you earn ` 130 and you need to
pay ` 101.16. In other words, you make a profit of ` 28.84 (` 130 − ` 101.16).
And, on the whole contract (which includes 50 Nifty), you will end up with
the profit of ` 1442.

Note While assessing options market using futures market, it assumed that

futures market is efficient. That is, it will converge to spot price at the maturity
of the contract.

4.3 DATA

4.3.1 Data Related TO Options Contracts, Spot Market, Futures

Market and Interest Rates
The PCP relationship has been tested on a data set containing data pertaining
to the four major categories, namely (i) data related to S&P CNX Nifty index
options contracts, (ii) data related to S&P CNX Nifty index futures, (iii) data
related to the underlying asset, i.e. the S&P CNX Nifty index and (iv) data on
the risk-free rate of return. The data relates to the first 6 years of trading in
Indian options market, i.e. from June 04, 2001 (the date on which index options
were launched in Indian derivatives market) to June 30, 2007.
Unlike the LBCs where the call and put options quotes are analysed
individually, the PCP requires pairs of the call and put options matched to
76 ● Derivative Markets in India

each other on the basis of three criteria—‘deal date’, ‘date of expiration’ and
‘strike price’. Therefore, the total number of quotes of call options (40,298) has
been matched with those of put options (35,172) based on the above-mentioned
three criteria, for all the 6 years under reference. The matching so carried out
resulted into 28,839 pairs of call and put options having identical ‘deal date’,
‘expiration date’ and ‘strike price’. Moreover, since the study also attempts
to test the PCFP, these pairs have further been matched to the data relating
to index futures based on the ‘deal date’ and ‘expiration date’. The matching
so carried out resulted into the same number of triplets (i.e. 28,839), as each
pair of options could find a futures contract having the same ‘deal date’ and
‘expiration date’.

4.3.2 Transaction Costs

The transaction costs applicable to trading member organization have been
used to test the PCP relationship and to identify the arbitrage opportunities/
mispricing signals, if any. Since the brokerage firms are privileged with the
least cost (transaction cost) structure in the market amongst all categories of
participants, it would be appropriate from the standpoint of the assessment
of market efficiency to use ‘transaction costs applicable to such organizations’
as a proxy for the transaction costs.
This has been done in view of the fact that such transaction cost would
facilitate in identifying those arbitrage opportunities that, otherwise, could
have remained unidentified, had we used some other proxy for the transaction
costs (e.g. the transaction costs applicable to retail investors). In sum, such
estimates of transaction costs will facilitate in assessing the true state of options
market efficiency in Indian derivatives market.

4.4 ANALYSIS AND EMPIRICAL RESULTS

The violations of the ‘put-all parity (PCP) relationship’ have been identified in
terms of underpricing and overpricing of put options relative to corresponding
call options with same contract specifications. A violation has been termed as
overpriced put option when the theoretical price of the put option (arrived
at based on PCP relationship using market price of the corresponding call
options) is found to be less than the market price of that put option. Similarly,
underpricing of put option has been registered when the market price of the
put option is less than the theoretical price of that put option. The violations so
identified have been analysed with respect to the specified levels of liquidity,
maturity and years under reference for both underpriced and overpriced put
options.
In the chapter, three levels of liquidity have been specified considering
the liquidity (number of contracts traded) of both call and put options with
same contract specifications. This has been done in view of the fact that the
 A Test of Put–Call Parity Relationship on S&P CNX Nifty Index Options Market ● 77

arbitrage strategy required to exploit the mispricing signals identified using

PCP relationship makes use of both call option and put option along with the
underlying asset (referred to as ‘triplet’ in literature). Therefore, the levels of
liquidity have been specified based on the liquidity of both call and put options
as it becomes important in ascertaining the exploitability of the observed
mispricing signals. These levels are, namely (i) thinly traded, where the
number of contracts traded per day for call (put) options are less than or equal
to 100, with corresponding put (call) options having any number of contracts
traded; (ii) moderately traded, where call (put) options are having 101–500
contracts traded per day along with the corresponding put (call) options having
more than 100 contracts traded per day and (iii) highly traded, where call as
well as put options having more than 500 contracts traded per day.
Similarly, four levels have been specified for the maturity of options
contracts. The specified levels of maturity are—(i) 0–7 Days to maturity; (ii) 8–30
Days to maturity; (iii) 31–60 Days to maturity and (iv) 61–90 Days to maturity.

4.4.1 Analysis of the Behavior of Violations from the PCP

Relationship
The summary of overall results regarding the violations of parity condition is
reported in Table 4.1. It can be deciphered from the results that a large number
of violations, which are nearly 96% of the observations analysed, have been
recorded before considering the transaction costs. However, the violations
have shown a sharp decline after reckoning the transaction costs that are
applicable to the most privileged category of investors, i.e. brokerage firms.
This is borne out by the fact that total number of violations have reduced
to 15,259 (2,224 + 13,035), which amounts to 52.91% of the total number of
pairs analysed. The violations identified after reckoning the transaction costs
constitute the basis of the study on account of their practical appeal and,
therefore, have been analysed subsequently to draw inferences about the
options market efficiency.
The frequency of violations, per-se, seems to be alarming as the violations
identified after reckoning the transaction costs applicable to the trading
member organizations represents more than half of the observations analysed.
Moreover, it is interesting to note that majority of the violations have been
recorded in terms of the relative overpricing of put options as it represents a
major chunk of total violations identified, i.e. 45.20% out of the total 52.91% of
the violations belong to the overpriced put options. However, the remaining
violations (i.e. 7.71% of the violations) connote the underpricing of put
options. Besides, the classification of violations in terms of the three levels of
liquidity demonstrates that a vast majority of violations in underpriced puts
belong to the thinly traded category, which can reasonably be designated as
unexploitable on account of the high bid-ask spread and high immediacy/
78 ● Derivative Markets in India

liquidity risk. In this respect, Ofek et al. (2004) aptly opined that the lack of
liquidity imposes another significant cost to arbitrageurs in terms of the bid-ask
spread, as it is higher for illiquid contracts and could turn out to be a major
constraint to arbitrageurs.
Likewise, the overpriced puts tread the same path as far as the frequency
of thinly traded violations is concerned, however, with the lower percentage
(68% of total violations in terms of overpriced puts) compared with that in
case of underpriced puts. Naturally, the percentage of violations belonging
to the relatively exploitable category, namely moderately and highly traded
category, is considerably higher for the overpriced put options compared
with that observed in the case of underpriced put options. That is, 32% of
the overpriced puts pertain to the ‘moderately and highly traded’ category
compared with 16% in the case of underpriced put options.
Table 4.1 Violations of the PCP Relationship and Liquidity Levels, June 2001–07
Underpriced Overpriced
Total
Particulars put options put options
(Percentage)
(Percentage) (Percentage)
Total number of observation analysed 28,839 28,839 28,839
Total number of violations observed 6413 21309 27,722
before transaction costs (22.24) (73.89) (96.13)
Total number of violations observed 2,224 13,035 15,259
after transaction costs (7.71) (45.20) (52.91)
Violations relating to the three specified levels of liquidity
1873 8881 10754
(a) Thinly traded options
(84) (68) (71)
304 2187 2491
(b) Moderately traded options
(14) (17) (16)
47 1967 2014
(c) Highly traded options
(2) (15) (13)
Total 2,224 13,035 15,259

The results on the frequency of violations suggest that majority of violation

have been recorded in terms of overpriced put options. However, it would be
appropriate to look at their magnitudes of profit for ensuring the exploitability
of such options. In view of this, the magnitudes of overpriced as well as
underpriced put options have been classified as per the specified categories
of liquidity. The analysis so made has been reported in Table 4.2. The results
clearly demonstrate that the overpriced put options offer higher magnitude of
profit along with the comparatively higher frequency of violations. It is evident
from the fact the average profit for the moderately and highly traded categories
in case of overpriced puts are ` 653 and ` 705 per contract, respectively; on
the other hand, these categories offer a considerably lower profit of ` 209 and
` 239, respectively, per contract in the case of underpriced put options. Based
on these observations, it may, prima-facie, be concluded that the overpriced
 A Test of Put–Call Parity Relationship on S&P CNX Nifty Index Options Market ● 79

put options are more likely to be exploited compared with underpriced put
options on account of higher magnitude of profits they offer.
Thought, it may be noted that the higher standard deviations both in the
cases of overpriced and underpriced put options (for both the categories) defy
the credibility of the mean as a correct statistical measure as these indicate
the presence of outliers in the data. Therefore, it would be reasonable to use
positional measures (i.e. median, percentiles, etc.) instead of the arithmetical
measures (i.e. mean). In view of this, the quartiles have been calculated for
the magnitude of profit and are reported in Table 4.2.
The results summarized in Table 4.2 clearly indicate that the violations
belonging to highly and moderately traded category in the underpriced put
options amount to just 351 (304 + 47) observations. In operational terms, such
violations in underpriced puts category can be designated as exploitable
since the bid-ask spread is expected to be relatively lower on account of the
higher trading in such violations. Moreover, the information contained in the
quartiles suggests that nearly one-fourth of the highly traded underpriced
puts (as indicated by the third quartile) can be designated as exploitable as
the third quartile offers a profit magnitude of more than ` 361. However, the
proportion of exploitable violations in moderately traded underpriced puts
category goes further down since the third quartile offers a starting profit of
less than ` 300. Further, in order to have a better idea about the percentage
of exploitable violations in this category, the 80th, 85th and 90th percentiles
have been calculated. The results suggest that nearly 20% of violations in
‘moderately traded underpriced put’ category can be reasonably designated
as exploitable because the 80th percentile reported a starting profit of ` 346
per contract.
In sum, it would be reasonable to conclude that a total number of 72 cases
(12 cases from the highly traded category and 60 from the moderately traded
underpriced puts) out of the total number of (2,224) violations identified in
terms of underpriced futures can be designated as exploitable. In other words,
the exploitable number of underpriced put options amounts to just 0.25% of
total pairs analysed. The rest of the violations in underpriced puts belong
to thinly traded category and amount to 84% of the total violations in this
category. Such violations can be termed as unexploitable on account of low
liquidity, which causes the bid-ask spread to be significantly higher.
Given the meagre percentage of exploitable violation in uderpriced puts
category, it would be reasonable to conclude that the put options are not
underpriced or call options are not relatively overpriced. This finding defies the
popular belief that call options, in general, are costlier than the put options.
In contrast, a significantly noticeable number of violations have been
observed in terms of overpriced put options. It is well manifested in the fact
that a total number of 4,154 (2,187 + 1,967) violations belonging to highly and
 80
●

Table 4.2 Liquidity-wise Descriptive Statistics for Violations of the PCP Condition for Underpriced and Overpriced Put Options in Indian
Derivative Markets in India

Securities Market, June 2001–07

Underpriced put options Overpriced put options
Liquidity Number of violations Magnitude of violations Number of violations Magnitude of violations
(Percentage) Mean S.D. Q1 Q2 Q3 (Percentage) Mean S.D. Q1 Q2 Q3
1,873 8,881
Thinly traded 563 966 89 237 611 789 1,056 196 483 1023
(84) (68)
304 2,187
Moderately traded 209 335 43 101 274 653 689 176 421 887
(14) (17)
47 1,967
Highly traded 239 230 58 169 361 705 630 238 539 997
(2) (15)
Total 2,224 508 905 76 211 534 13,035 754 950 197 480 999
In the table, S.D., Q1, Q2 and Q3 denote standard deviation, first quartile, second quartile (i.e. median) and third quartile respectively.
 A Test of Put–Call Parity Relationship on S&P CNX Nifty Index Options Market ● 81

moderately traded category have been identified. These violations amount

to 32% of total number of violations in overpriced put options. Since the
bid-ask spread is expected to be considerably low for such options, it would
be reasonable to designate these violations as exploitable. In addition, the
finding is revealing that nearly 50% of such mispricing opportunities seem
exploitable as the second quartiles report a starting profit of ` 421 and `539
per lot for moderately and thinly traded categories, respectively. In sum, it
may be appropriate to conclude that a significantly alarming/higher number
of violations in terms of overpriced put options can be labelled as exploitable
in light of the magnitude of profits they offer. It is eloquently borne out by
the fact that the exploitable violations in this category turn out to be more
than 2000 (half of the violation in moderately and traded category), which
represent nearly 7% of the total observations anlysed.
However, the remaining violations in terms of overpriced put options
remained unexploited because of lower liquidity they possessed (i.e. thinly
traded options). Such violations accounts for the more than two-third
of the violations in this category. In this respect, commenting upon the
unexploitability of such options, Kamara and Miller Jr (1995) observed that
such violations essentially mirror the liquidity or immediacy risk deriving
from the possibility of unfavourable price shifts between the time when the
decision to build strategies is made and the actual execution of the orders. And,
it further cautioned that the liquidity risk is particularly high in the case of
index options since arbitrage portfolios often require trading the stock basket
which constitutes the index.
In view of the above-mentioned findings, it may aptly be concluded that
the mispricing identified using PCP analysis revealed the relative overpricing
of put options to the corresponding call options, as a significantly noticeable
number of violations pertaining to the overpriced put category could be
designated as exploitable.
Besides, the violations so observed in terms of overpricing and underpricing
of put options have further been analysed with respect to the maturity time
they had. This has been attempted with intent to understand the clustering of
violations as it helps to draw some meaningful inferences for the betterment of
the market as well as helps to find more reasons for the existence of violations
in the market. Moreover, the classification so made has further been sub-
classified within each category as per the three specified levels of liquidity.
The sub-classification so made aims at ensuring the exploitability of the
violations contained in the four specified levels of maturity and, thus, facilitates
in assessing the state of efficiency across the different levels of maturity. The
results in this respect have been reported in Table 4.3.
The results summarized in the table indicate that majority of violations
both in the case of underpriced and overpriced put options are clustered in
 82
●

Table 4.3 Maturity-wise Descriptive Statistics for Violations of the PCP Condition for Underpriced and Overpriced Put Options in Indian
Securities Market, June 2001–07
Underpriced put options Overpriced put options
Days to
Liquidity Number of Magnitude of violations Number of Magnitude of violations
maturity
violations Mean S.D. Q1 Q2 Q3 violations Mean S.D. Q1 Q2 Q3
Thinly traded 561 585 926 93 263 657 879 646 1386 106 283 666
Derivative Markets in India

‘0–7’ Moderately traded 94 285 515 60 116 353 147 357 504 83 208 379
Days Highly traded 24 245 254 41 171 355 107 462 620 109 244 414
Overall 679 532 872 84 231 604 1133 591 1253 104 265 590
Thinly traded 925 556 985 88 230 604 4689 602 847 166 376 722
‘8–30’ Moderately traded 190 171 171 41 97 250 1229 453 511 139 309 598
Days Highly traded 23 233 207 60 167 373 1351 571 488 212 466 792
Overall 1138 485 903 71 195 503 7269 571 745 167 379 720
Thinly traded 366 867 867 82 211 503 2734 964 1031 304 735 1330
‘31–60’ Moderately traded 20 396 396 11 60 207 716 916 719 334 789 1340
Days Highly traded Nil Nil Nil Nil Nil Nil 437 989 673 474 931 1418
Overall 386 484 851 68 199 497 3887 958 946 323 770 1345
Thinly traded 21 1413 2007 411 872 1766 579 1695 1445 759 1489 2255
‘61–90’ Moderately traded Nil Nil Nil Nil Nil Nil 95 1712 994 861 1702 2422
Days Highly traded Nil Nil Nil Nil Nil Nil 72 1841 945 1178 1811 2560
Overall 21 1413 2007 411 872 1766 746 1711 1353 782 1544 2348
In the table, S.D., Q1, Q2 and Q3 denote standard deviation, first quartile, second quartile (i.e. median) and third quartile, respectively.
 A Test of Put–Call Parity Relationship on S&P CNX Nifty Index Options Market ● 83

the 0–60 days to maturity category, i.e. the first three levels of maturity, viz.
0–7, 8–30 and 31–60 days to maturity. As already revealed in the analysis of
violations in relation to different levels of liquidity, the number of exploitable
violations in terms of the underpricing of puts seems to be negligible across
all the categories of maturity. It is empirically evident from the fact that the
magnitudes of the violations pertaining to highly and moderately liquid
categories are very low. The very first category in the 0–7 days to maturity have
a total number of 118 violations including cases of both moderately and highly
traded categories. Moreover, only the third quartiles of such violations offer a
starting amount of magnitudes (viz. ` 353 and ` 355, respectively) that can be
designated as worth exploiting in the presence of bid-ask spread. Therefore, it
may reasonably be deciphered that only one-fourth of such violations (nearly
30 cases) are exploitable.
Similarly, for category 8–30 days to maturity of underpriced put options
have a total number of 213 violations aggregately in moderately and highly
liquid categories. However, if we look at the exploitability of such violations,
the number must certainly go down as only the third quartile of highly traded
category offers a potentially exploitable starting profit of ` 373. Besides, the
third quartile in case of moderately traded category offers a lower profit
margin, which per-se, suggests that even less than one-fourth of such violations
may be exploitable. Therefore, to have precise idea about the number of
exploitable cases in this category, 80th, 85th and 90th percentile were calculated.
The 80th percentile for such options suggested a starting profit of ` 317, which
indicates that 20% of the cases in this category can be designated as exploitable.
In sum, a total number of 44 cases of violations in 8–30 days to maturity can
be designated as exploitable, which is again a meagre number.
Moreover, the last two categories of maturity for the underpriced put
options revealed negligible to zero exploitable cases as there are no cases in
the highly liquid category for both the maturity levels. Besides, the moderately
traded category have negligible cases in the case of violations having 31–60
days to maturity and no cases for the last category.
In sum, it may be inferred from the above discussion that the number of
exploitable violations remained negligible for the first two levels of maturity.
Moreover, such violations remained virtually absent for the last two categories.
A vast majority of such cases across all the levels of maturity remained
unexploited in view of their profit magnitudes and the immediacy risk.
In sharp contrast to the underpriced put options, the overpriced put
options seem to offer noticeable number of exploitable violations across all the
categories of maturity. Moreover, such violations predominately appeared in
the second and third level of maturity, i.e. 8–30 and 31–60 days to maturity.
It is empirically corroborated from the fact that the magnitude of profits in
case of overpriced put options are substantially higher compared with those
84 ● Derivative Markets in India

in the case of underpriced puts across all the categories. In the 0–7 days to
maturity category, one-fourth of the total cases in moderately and highly
traded category (nearly 60 cases); based on the magnitudes of profit they offer
can be referred to as exploitable. The finding is revealing as long as the number
of exploitable violations in the next two maturity levels are concerned. This is
evident from the fact that half of the violations in moderately as well as highly
traded category for the maturity level 8–30 days to maturity (which amount to
more than 1250 cases) could be designated as exploitable based on the profit
magnitude they offer. Moreover, nearly an equal number of violations for the
next maturity level, i.e. 31–60 days to maturity can be designated as exploitable,
as all such violations belonging to moderately and highly traded category
offered attractive profit potentials. Besides, unlike the underpriced put options,
the overpriced put options experienced the existence of a significant number
of violations in the 61–90 days to maturity category as well.
In sum, it may be concluded that the overpriced put options had a noticeable
number of exploitable violations across all the four categories of time to
maturity unlike the underpriced puts, where negligible number of violations
were recorded across all the four levels. The violations predominantly occurred
in 8–30 and 31–60 days to maturity categories.

4.4.2 Statistical Significance of the Difference in the Magnitude of

Violations
In view of the above findings, it can be observed that there seems to be a
difference in the mean magnitudes of violations among the specified levels
of liquidity for underpriced as well as overpriced put options. Therefore, to
validate the difference statistically, Kruskal–Wallis (H-statistics) test, which
is a non-parametric substitute for the one-way ANOVA, has been applied. It
has been done in view of the fact that the main assumption of ANOVA, i.e. all
the samples have been drawn from a normally distributed population, could
not be validated for the data under consideration. The normality assumption
for all the samples has been tested using ‘one-sample Kolmogorov–Smirnov
statistics’. The results reported in Table 4.4 reveal severe departure from the
normality. And, therefore, it would not be appropriate to apply ANOVA for
testing the differences of magnitudes across all the three levels of liquidity.
Therefore, the differences have been analysed using a non-parametric statistics,
which does not require the data to follow any specified distribution.
In addition to this, Dunn’s multiple comparison statistics has been used for
post-hoc analysis of all possible pairs in the analysis. The results of H statistics
and Dunn’s test for the differences across the specified levels of liquidity,
respectively, are summarized in Tables 4.5 (a) and 4.5 (b).
The significance value given in Table 4.5 (a) clearly indicates that the
difference among the specified levels of liquidity is highly significant, as it has
 A Test of Put–Call Parity Relationship on S&P CNX Nifty Index Options Market ● 85

been found to be significant even at 1% level of significance for underpriced as

well as overpriced put options. As a result of significant results of H statistics,
the post-hoc diagnosis, i.e. Dunn’s multiple comparison test, has been carried out
to find the pair-wise differences.
Table 4.4 Summary of the One-Sample Kolmogorov–Smirnov Statistics
to Assess the Normality
Underpriced put Overpriced put
options options
Number of observations 2224 13035
Normal parameters Mean 507.739 753.657
(a), (b) Std. deviation 905.019 949.9734
Absolute 0.288 0.214
Most extreme Positive 0.242 0.162
differences
Negative – 0.288 – 0.214
Kolmogorov–Smirnov Z 13.562 24.426
Asymp. Sig. (2-tailed) 0.000 0.000
(a) Test distribution is Normal; (b) Calculated from data.

The results of the post-hoc analysis in the case of call options summarized
in Table 4.5(b) signify that average magnitude of violations for the thinly
traded options is significantly different from that for the moderately traded
but not significaltly different from the highly traded category in the case of
underpriced put options. Besides, there has been no significant difference
between the magnitudes of moderately and highly traded category as well.
In contrast, all the pairs were found to have significantly different magnitude
of profits except for the one pair, i.e. Thinly vs. Highly traded pair in the case
of overpriced put options.
In operational terms, the results imply that the average magnitude of
violations for exploitable options contracts is significantly different from those
for options contracts that can be designated as unexploitable. Moreover, it may
be noted that in both underpriced and overpriced put options, the magnitude
of highly traded options is not significantly different from those of thinly
traded options, but the latter could not be designated as exploitable in light
of the higher bid-ask spread and immediacy risk.
The finding, per-se, indicates that the profit potential of exploitable
opportunities has been quite high, especially for overpriced put options, as
the profit for highly traded options has been found to be higher than that for
thinly traded category; however, the increase was not statistically significant
at 5% level of significance. Moreover, the profit of highly traded category for
overpriced put options has been considerably higher than that of moderately
traded category. It, therefore, indicates that the exploitable opportunities in
case of overpriced put options offered quite high amount of profit.
 86
●

Table 4.5 (a) Kruskal–Wallis (H-statistics) Test for the Differences among the Violations across the Specified Levels of Liquidity for Underpriced and
Overpriced Put Options, June 2001–07
Underpriced put options Overpriced put options
Liquidity Rank Test statistics (a), (b) Rank Test statistics (a), (b)
N Mean rank Chi-square df Sig. N Mean rank Chi-square df Sig.
Thinly liquid 1873 1164.98 8881 6562.72
Derivative Markets in India

Moderately liquid 304 815.46 80.86 2 0.000 2187 6135.00 30.873 2 0.000
Highly liquid 47 942.35 1967 6741.90
(a) Kruskal Wallis Test; (b) Grouping Variable—Liquidity

Table 4.5 (b) Dunn’s Test for the Multiple Comparisons amongst the Specified Levels of Liquidity for Overpriced as well as Underpriced Put Options,
June 2001–07
Call options Call options
Dunn’s multiple
comparison test Difference in Significant Difference in Significant
Summary Summary
rank sum (P < 0.05) rank sum (P < 0.05)
Thinly traded
vs. 349.5 – Yes 427.7 – Yes
Moderately traded
Thinly traded
vs. 222.6 Ns No −179.2 Ns No
Highly traded
Moderately traded vs.
−126.9 Ns No −606.9 – Yes
Highly traded
 A Test of Put–Call Parity Relationship on S&P CNX Nifty Index Options Market ● 87

4.4.3 Learning Experience in the Indian Options Market

This part of the chapter has attempted to examine the learning behavior of the
investors in the Indian options markets over the period under reference. For
the purpose, it has been hypothesized that number of violations in the market
should go down over the years as the investors are expected to be familiar
with the market in due course of time. Precisely, the hypothesis warrants
improvement in the market efficiency over the years as the investors are expected
to behave more rationally in pricing the options contracts. The examination of
the violations over the years under reference is in line with Mittnik and Rieken
(2000), a study carried out in the context of German index options market.
The results summarized in Table 4.6 and Fig. 4.1 clearly indicate that the
percentage of mispricing signals relative to the total number of observations
analysed has shown a considerable decline over the years for ‘underpricing
of put options’. The decrease in the frequency of such mispricing signals
demonstrates that the underpricing of put options (as suggested by PCP
relationship) has improved significantly over the years. In contrast, the
observed overpricing of put options shows an almost constant to increasing
trend in the percentages of mispricing signals over the years (Table 4.6 and
Fig. 4.2). The observed persistence in these mispricing signals evidently rules
out the improvement in the efficiency of the options market (as warranted by
the learning hypothesis) over the years.

20.00%

15.00%

10.00%

5.00%

0.00%
2001–02 2002–03 2003–04 2004–05 2005–06 2006–07

Figure 4.1 Percentage of relatively underpriced put options in relation to the

number of pairs analysed in the year over the sample period

In addition, the magnitude of the violations across the years has been
examined with intent to ensure exploitability of violations over the years.
The results have been summarized in Table 4.7. It is satisfying to note that
the frequency of exploitable violations in the case of underpriced put options
remained negligible across all the years except the year 2003–04. Moreover, the
year 2003–04 accounts for more than two-fifth of the total violations occurred in
underpriced put category. For the rest of five years, the frequency of exploitable
violations remained negligible in view of the magnitude of profit they offered
for the exploitable category, i.e. the moderately and thinly traded category.
 88
●
Derivative Markets in India

Table 4.6 Year-wise Frequencies of the Violations of PCP Relationship, June 2001–07
Total violations Put options underpriced Put options overpriced
Total number of
Year No. of Percentage of No. of Percentage of No. of Percentage of
pairs analysed
violations violations (%) violations violations (%) violations violations (%)
2001–02 2226 1347 60.51 303 13.61 1044 46.90
2002–03 2922 1509 51.64 480 16.43 1029 35.22
2003–04 5296 3020 57.02 964 18.20 2056 38.82
2004–05 5390 2414 44.79 127 2.36 2287 42.43
2005–06 6368 3690 57.95 222 3.49 3468 54.46
2006–07 6637 3279 49.40 128 1.93 3151 47.48
Total 28839 15259 52.91 2224 7.71 13035 45.20
 A Test of Put–Call Parity Relationship on S&P CNX Nifty Index Options Market ● 89

60.00%

50.00%

40.00%

30.00%

20.00%

10.00%

0.00%
2001–02 2002–03 2003–04 2004–05 2005–06 2006–07

Figure 4.2 Percentage of relatively overpriced put options in relation to the

number of pairs analysed in the year over the sample period

In sum, it would be appropriate to state that the number of exploitable

violations remained negligible for underpriced put options for all the years
except the year 2003–04. However, the year 2003–04 revealed noticeable
number of exploitable violations as the third quartiles for moderately as well
as highly traded category offered a reasonable starting profit of ` 336 and
346, respectively.
In contrast, a significantly noticeable number of exploitable violations have
been recorded in case of overpriced put option over the years. Moreover, the
finding is revealing as the number of exploitable violations in this category
has shown an increasing trend over the years. Empirically, the increase in the
number of exploitable violation is evident from the results summarized in
Table 4.7. The results demonstrate that the number of exploitable violations
(i.e. violations contained in the moderately and highly traded categories with
the reasonable profit magnitude) have been nearly 35, 150 and 180 in the
first three years of trading. Notably, there has been an alarming increase in
the frequency of exploitable violations in the next three years of trading, i.e.
2004–05 to 2006–07.
The number of exploitable violations recorded in the year 2004-05 showed
an increase of more than 100% compared with last years, as the number of
cases increased from 150 to 340. Likewise, the next year registered an increase
of more than 50% compared with the year 2004–05. Empirically, the last year
under reference, i.e. 2006–07 has shown a dramatic increase in the number of
violations, as it registered a total number of more than 1200 cases, which can
be designated as exploitable in light of the profit magnitude they offered.
In sum, the violations in terms of overpriced puts have registered a
significant increase in the number of exploitable violations over the years
compared with underpriced put options where the number of exploitable
violations remained negligible in all the years except the year 2003–04. This
finding has further corroborated the trend of violations revealed initially by
the percentage of violations vis-à-vis the number of pairs analysed (reported
 Table 4.7 Year-wise Descriptive Statistics for the Magnitude of the PCP Condition for Underpriced and Overpriced Put Options in Indian 90
Securities Market, June 2001–07
●

Underpriced put options Overpriced put options

Years Liquidity Number of Magnitude of violations (INR) Number of Magnitude of violations (INR)
violations Mean S.D. Q1 Q2 Q3 violations Mean S.D. Q1 Q2 Q3
Thinly traded 280 250 446 45 142 268 971 379 321 139 289 551
Moderately traded 23 95 88 39 45 154 73 427 245 238 394 651
2001–02
Highly traded Nil Nil Nil Nil Nil Nil Nil Nil Nil Nil Nil Nil
Overall 303 246 433 45 138 245 1044 382 316 140 291 554
Derivative Markets in India

Thinly traded 394 188 218 54 106 250 797 254 237 88 193 m
Moderately traded 82 98 107 41 61 97 219 186 160 51 147 292
2002–03
Highly traded 4 316 179 226 405 405 13 207 167 54 109 294
Overall 480 174 206 48 95 241 1029 239 223 58 188 343
Thinly traded 744 529 737 111 300 652 1569 604 677 168 387 817
Moderately traded 187 246 379 58 153 336 363 415 410 130 273 528
2003–04
Highly traded 33 219 238 35 102 346 124 480 363 171 371 792
Overall 964 464 681 76 250 560 2056 563 627 163 370 776
Thinly traded 125 604 824 110 334 754 1599 587 631 196 406 721
Moderately traded 1 80 Nil Nil Nil Nil 361 453 405 153 348 616
2004–05
Highly traded 1 830 Nil Nil Nil Nil 327 531 423 208 422 696
Overall 127 602 819 109 334 761 2287 558 576 196 401 706
Thinly traded 206 1315 1662 295 728 1766 2310 1042 1257 298 739 1375
Moderately traded 9 551 517 129 353 853 556 833 809 271 577 1120
2005–06
Highly traded 7 222 120 161 220 246 602 659 690 172 447 870
Overall 222 1250 1622 269 687 1651 3468 942 1124 262 649 1296
Thinly traded 121 1383 1678 338 820 1735 1635 1313 1494 417 970 1733
Moderately traded 5 518 477 297 431 431 615 942 797 278 770 1394
2006–07
Highly traded 2 193 206 47 193 338 901 836 653 344 724 1159
Overall 128 1330 1685 45 142 268 3151 1104 1205 374 840 1496
In the table, S.D., Q1, Q2 and Q3 denote standard deviation, first quartile, second quartile (i.e. median) and third quartile, respectively.
 A Test of Put–Call Parity Relationship on S&P CNX Nifty Index Options Market ● 91

in Table 4.6, Figs 4.1 and 4.2). In short, a vast majority of exploitable violations
appeared in the case of overpriced put options and, therefore, failed to show
the warranted improvement in the market efficiency unlike the underpriced
put options, which were found consistent with the learning hypothesis.

4.4.4 Comparison of Underpriced and Overpriced Put Options

The analysis of PCP has clearly demonstrated that the number of exploitable
violations in the form of overpriced put options have consistently been higher
compared with those in the case of underpriced put options. This finding is
in line with a number of studies varied out across the globe, e.g. Goh and
Allen (1984) and Mittnik and Reiken (2000b). In view of this, an attempt has
been made to trace the reason behind the lopsided tendency of violations. The
phenomena can be offered a plausible explanation if we look at the arbitrage
strategies required to correct such pricing anomalies. The strategy required
to exploit profit opportunities emanating from the relative underpricing of
put options entails a long position in the underlying asset. However, the
exploitability of mispricing signals in case of ‘overpriced put options’ requires
taking a short-position in the underlying asset.
Commenting upon the difficulty in shorting an index, Mittnik and Reiken
(2000b) opined that the short position in the assets, especially when the asset is
an index, is more difficult compared with a long position and requires higher
transaction costs. In addition to this, the short-selling constraint on financial
assets has been reported as the major reason for such a development in various
financial markets across the world. In view of the above, the observed relative
overpricing of the put options in Indian options market can be attributed to
the market-microstructure which did not allow to short-sell the financial assets
during the period under reference. As already mentioned, an arbitrageur
needs to sell the underlying asset short to exploit an arbitrage opportunity in
the form of relative-overpricing of puts, the short-selling constraint in Indian
options market caused the relative overpricing of puts to persist.
These findings are consistent with those of Evnine and Rudd (1985),
Finucane (1991), Brown and Easton (1992), Bharadwaj and Wiggins (2001)
and Ofek et al. (2004) for different markets.

4.4.5 Violations Using Futures Market

In addition, an attempt has been made to identify the violations using put-call
futures parity condition, i.e. test of parity condition using futures prices of
the underlying assets instead of the spot values. This has been done in view
of the facts that
(i) the short-selling of the financial assets was not permitted in the Indian
securities market during the period under reference and futures market
easily provides this facility;
92 ● Derivative Markets in India

(ii) in case of options having index as the underlying asset, it becomes very
difficult to short-sell/purchase of the index as all the basket of shares
which constitute the index need to be sold/purchased in the proportion
in which these have been included in the index and
(iii) relatively less transactions costs associated with trading in futures
market compared with the spot market. The results on the violations
of put-call-futures condition are reported in Table 4.8.
The results reported in Table 4.8 reveal that the number of violations of
PCFP remained quite high, predominantly, in terms of overpricing of put
options, as a vast majority of violations belong to this category. Notably, unlike
identifying violations of PCP, short-selling constraint, prima-facie, could not be
a correct explanation to the higher frequency of violations in call options since
to exploit the arbitrage opportunities using futures market, the arbitrageur
does not have to short the stock basket, rather he needs to sell the futures that
is easily possible.
Moreover, another notable finding is that the violations of the parity
condition remained nearly at the same level as revealed by the analysis using
the spot market (Table 4.8). It is borne out by the fact that such violations
have shown a negligible decline from 13,035 to 13,016. Notwithstanding the
trend in overpriced put options, a substantial increase has been recorded in
violations labelled as underpriced put options. The violations in terms of the
under pricing of put has been as high as four times of those in the case of
parity analysis using spot market values. As such, violations have increase
from 2224 to a whopping 9612.
An intuitive explanation to such a development could be traced to the fact
that the transaction costs for the trading in the futures market are substantially
lower compared with those for spot market transactions. It seems to be a
plausible reasoning for the observed increase in the number of underpriced
put options. However, there should have been a corroborating development
(increase) in the overpricing of puts as well to make this explanation tenable.
Empirically, this is not the case as the violations in terms of the overpricing
of put options have shown a negligible decline and, therefore, discards this
explanation to above-mentioned development in the frequency of violations.
Moreover, the frequency of violations without considering the transaction
costs (both in the case of PCP and PCFP, Table 4.8) offers strong corroborating
evidences to the above-mentioned finding, i.e. the transaction costs cannot be
offered a major reason for the observed decline (increase) in the number of
overpriced (underpriced) puts options.
Notably, another more plausible explanation in terms of the ‘underpricing
of futures contracts’ can be offered for the observed change in the trend of
violation while using futures market instead of the spot market. It is for this
reason that there has been an increase (decrease) in the number of underpriced
 Table 4.8 Violations of the PCP Relationship Using Spot Values and Futures Prices, June 2001–07: A Comparative Analysis
Using spot values Using futures prices
t

Particulars Underpriced Overpriced Underpriced Overpriced

Total Total
put options put options put options put options
(Percentage) (Percentage)
(Percentage) (Percentage) (Percentage) (Percentage)
Total number of observation
28,839 28,839 28,839 28,839 28,839 28,839
analysed
Total number of violations
6,413 21,309 27,722 12,176 16,600 28,776
observed before transaction
(22.24) (73.89) (96.13) (42.22) (57.56) (99.78)
costs
Total number of violations
2,224 13,035 15,259 9,612 13016 22,628
observed after transaction
(7.71) (45.20) (52.91) (33.33) (45.13) (78.46)
costs
A Test of Put–Call Parity Relationship on S&P CNX Nifty Index Options Market
●
93
94 ● Derivative Markets in India

(overpriced) put options. The effect of underprricing of futures on the relative

overpricing/underpricing of put options can be understood with the help of
Eq. (4.8).
Pt = Ct – e– r(T – t) ( Ft – K)

Æ
Æ
(4.8)
The equation demonstrates that the underpricing of futures is liable to cause
an increase in the theoretical price of the put options, given the market price
of the call option. And, therefore, the put options that have been found to be
correctly priced or overpriced in the test of PCP using spot prices are likely
to be converted into the underpriced or correctly priced put options given the
magnitude of increase in the theoretical price of the put option. The process of
conversion of correctly priced and overpriced put options into the underpriced/
correctly priced ones has been well documented in the number of violations
without considering the transaction costs, as reported in Table 4.8. The results
show that there has been an increase in the number of underpriced puts from
6,413 to 12,176 on account of the use of futures market. Similarly, there has
been a decrease in the number of overpriced puts from 21,309 to 16,600 on
account of the use of futures market instead of the spot market. In sum, the
use of futures market, essentially on account of the underpricing of futures
contracts, has registered an increase in the underpriced puts by 5,763 and,
at the same time, a decrease in the overpriced put options by 4,709. In other
words, the underpricing of futures contracts has converted 4,709 overpriced
put options into the underpriced put options. Likewise, a total number of 1054
put options contracts, which were found to be correctly priced (in the test of
PCP using spot price), have been designated as underpriced on account of
the underpricing of futures contracts. In sum, the decrease (increase) in the
number of overpriced (underpriced) can appropriately be designated to the
underpricing of futures contracts in the Indian derivatives market, for the
period under reference.
The observed underpricing of futures contracts may be traced to the short-
selling constraint in the Indian securities market during the period under
reference. The non-availability of the short-selling facility can be designated
as the primary reason for the underpricing of the futures contracts because
of fact that the arbitrage strategy needed to correct such pricing anomalies
required short-selling of the underlying asset. In short, the underpricing of
the futures can, therefore, be attributed to the fact that short-selling has been
banned during the period under reference in Indian securities market.
Thus, it may be concluded that the futures market has failed to provide a
good substitute for shorting the assets in the absence of short-selling facility
in the Indian securities market during the period under reference. Therefore,
the futures market could not succeed in restoring equilibrium in the options
market.
In view of this, it is reasonable to conclude that the indirect impact of the
short-selling constraints on the efficiency of the options market on account of
 A Test of Put–Call Parity Relationship on S&P CNX Nifty Index Options Market ● 95

the interrelationship of the index options and index futures market has been
one of the major reasons amongst others (e.g. liquidity) for the existence of
mispricing signals in the Indian options market. In short, the impact of short-
selling constraints cannot be ignored even if the violations are identified
using futures contracts, as the efficiency of futures market does impact the
efficiency of options market, which in turn, can be ensured when short-selling
is allowed.

4.5 CONCLUDING OBSERVATIONS

This chapter attempted to examine the PCP condition for the (European type)
S&P CNX Nifty index options prices using the spot as well as futures prices
on the same index in the Indian securities market. The results are revealing as,
empirically, more than half of the pairs analysed violated the PCP condition
after accounting for the transaction costs. Out of the total number of violations
observed, a vast majority of violations were identified in terms of relative
overpricing of put options; a relatively meagre number of violations showed
underpricing of put options. The number of violations is alarming in view
of the fact that the frequency of violations identified in Indian market is
substantially higher compared with those found in developed economies. For
example, Wagner et al. (1996), a study in the US context, reported that 21.1%
of the pairs violated PCP in S&P index options market. However, it may be
noted that such violations demonstrated a sharp decline when seen in terms
of required liquidity to designate them exploitable. The lack of adequate
liquidity indicates higher bid-ask spread as well as immediacy risk/liquidity
risk. The finding that the options are underpriced is consistent with that of
Varma (2002), a study carried out in Indian context.
Besides, the concentration of violations both for underpriced as well as
overpriced put options have been in the overall category of 0–60 days to
maturity, which primarily appeared in the category 8–30 and 31–60 days to
maturity.
The study is equally revealing as far as the behaviour of the investors
dealing with the options market in India is concerned. It has been observed
that the number of violations in call options market has been persistent instead
of showing a warranted declining trend. In other words, it implies that the
irrationalities in the behavior of investors, particularly in call options market,
could not improve significantly over the years. However, it is satisfying to
note that the put options market has behaved in a way that is consistent with
the learning hypothesis, i.e. the number of violations has reduced with the
passage of time. Thus, the findings indicate that the put options market is
emerging to be more efficient vis-à-vis the call options market.
Another notable finding of the study is that the futures market could not
provide a good substitute for the short-selling facility in the Indian securities
96 ● Derivative Markets in India

market. This is eloquently borne out by the fact that the number of violations
in terms of relative overpricing of put options remained nearly same even
when the futures market has been used to identify the violations. Moreover,
the no. of violations in terms of underpricing of put options have shown a
whooping increase when identified using futures market. The development
so observed can be traced to the fact that the futures themselves remained
underpriced in the absence of short-selling facility in the market and, therefore,
registered a negligible decrease in overpriced put options and an increase
for the underpriced put options. In short, it may be appropriate to conclude
that in the absence of short-selling facility in Indian securities market for the
time period under reference, the futures themselves traded away from the
equilibrium prices. It is for this reason that the futures market failed to restore
equilibrium in options market.
In sum, it is reasonable to conclude that majority of violations identified
using PCP relationship could not be exploited on account of the existing market
microstructure in India during the period under reference (i.e. short-selling
constraint that caused underpricing in futures to persist). Moreover, the futures
market failed to restore efficiency in options market due to its own inefficiency
in terms of underpricing for the period under reference because of the lack of
short-selling facility during the same period. Besides, the dearth of liquidity
in the options market appears to be another major constraint to arbitrageurs
because a vast majority of violations occurred in the thinly traded category.
 Testing the Expectations Hypothesis on the Term Structure of Volatilities Implied by........... ● 97
5 Testing the Expectations
Hypothesis on the Term
C H A P T E R

Structure of Volatilities
Implied by S&P CNX Nifty
Index Options

5.1 INTRODUCTION
Term structure of implied volatilities connotes the relationship of implied
volatility with time. The pattern of implied volatility across time to expiration
is known as the term structure of implied volatility, and the pattern across
strike prices is referred to as the volatility smile (Dupire, 1994; Derman and
Kani, 1994 and Rubinstein, 1994). Term structures are helpful in pricing the
options for which the volatility of the underlying asset can be assumed to be
a deterministic function of time. It has been well documented in the available
literature that different restrictions have been imposed on the term structure
of implied volatility, considering the nature of underlying asset’s volatility.
In this chapter, a restriction on the term structure of implied volatility,
assuming rational expectations to hold, will be tested in line with the study
done by Stein (1989), Diz and Finucane (1993), Heynen et al.(1994), Campa and
Chang (1995) and Takezaba and Shiraishi (1998). The restriction is arrived at
assuming mean-reversion in implied volatility. The mean-reversion in implied
volatility connotes that, in long run, it will return to its long-term average,
which is assumed to be constant. Before testing the restriction on the term
structure, the mean-reversion property of the implied volatility has been
validated. In short, this chapter addresses two empirical questions, viz. (a)
whether the implied volatilities, in the case of short-dated as well as long-dated
options, are mean reverting and (b) whether the volatilities implied by the
long-dated options are consistent with the future volatilities estimated based
on the volatilities implied by corresponding short-dated options, as warranted
by rational expectations to hold. This has been operationalized by measuring
the empirical elasticity coefficients, which reveal observed or empirical causal
relationship between the volatilities implied by short-dated and long-dated
options, and comparing it with the theoretical elasticity coefficient derived
from the theoretical restriction on the implied volatility.
98 ● Derivative Markets in India

Since derivatives market, in general, and options market, in particular, is

in nascent stage in India, the chapter aims at diagnosing the inefficiencies
in pricing the index options. The inefficiencies has been diagnosed through
analysis of implied volatility, as volatility is the only unobservable variable
in valuation of the index option contracts while using the theoretical formula
suggested by Black and Scholes (1973).
The rest of the chapter has been organized in four sections. Section 5.2
discusses the methodology, which is divided in two sub-sections, namely 5.2.1
and 5.2.2. The sub-section 5.2.1 deals with the statistical techniques used to test
the mean-reversion in the implied volatilities, while sub-section 5.2.2 describes
the restriction on the term structure of implied volatilities to test the rational
expectations hypothesis (REH). The data and calculation of implied volatility
have been discussed in sub-sections 5.3.1 and 5.3.2, respectively. Section 5.4
presents the results relating to the mean-reversion and rational expectations.
The concluding observations are contained in the Section 5.5.

5.2 EXPECTATIONS HYPOTHESIS ON THE TERM STRUCTURE OF

IMPLIED VOLATILITIES

5.2.1 Mean-Reversion in Implied Volatilities

To test whether implied volatilities are mean reverting or not, we essentially
have to test the stationarity of the series of implied volatilities. A time series
is said to be stationary provided its mean, variance and autocovariance (at
different lags) remain the same irrespective of the point of time at which the
measurement is made. Such a stationary series is expected to return to its mean
in long run, technically termed as mean-reversion, Gujrati (2004).
In order to test the stationarity of a series, various techniques have
been developed. The two techniques which have been used widely are (1)
correlograms and (2) unit root tests. Of these, unit root test is considered more
useful statistical technique. The two tests mentioned are being discussed in
detail and will be applied to test the stationarity of the implied volatilities.
The correlogram is a graphical technique where autocorrelation and partial
autocorrelation coefficients are plotted against their respective lags. The graphs
so created are popularly known as autocorrelation functions (ACF) and partial
autocorrelation functions (PACF). In correlograms, ACFk, which specifies
autocorrelation at lag k, is nothing but the simple correlation between Yt and
Yt + k, where Y is a time series being analysed. Likewise PACFk, which connotes
partial autocorrelation at lag k, is the simple correlation between the Yt and
Yt + k minus that part explained linearly by intermediate lags.
Based on the examination of the patterns of ACF and PACF, it can be
decided whether a series is stationary or not. In addition to this, values of the
 Testing the Expectations Hypothesis on the Term Structure of Volatilities Implied by........... ● 99

Ljung-Box statistics, also referred to as Q-statistics, help us to know the lag

length for which the values of ACF and PACF are statistically significant. It
is very important to decide the lag length for which ACFs and PACFs should
be plotted to test the stationarity of the series. Empirically, it can be decided
by resorting to the thumb rule, which suggests that the ACFs and PACFs
should be computed up to one-forth to one-third of the length of the series
being analysed (Gujrati, 2004).
On account of the subjectivity involved in correlograms, it is better to use
a more objective technique called unit root test, which is a very popular tool
to test the stationarity. In the unit root test, we try to find out whether a series
contains a unit root or not. If a series contains a unit root, such a series is called
as a non-stationary series and vice versa. The starting point for the unit root
test is the unit root process, which is given in the Eq. (5.1).
Yt = aYt – 1 + ut (5.1)
where the value of a ranges between −1 and +1, and ut is a white noise error
term. A variable is referred to as a white noise when it has zero mean, constant
variance and is serially uncorrelated.
In Eq. (5.1), if the value of a = 1 (this is the case of unit root), the series Yt
becomes a non-stationary series. This is achieved by regressing Yt on its lag
Yt – 1 and then testing if the estimated value of a is statistically equal to 1 or not.
In practice, Eq. (5.1) is not tested directly but it is manipulated by subtracting
Yt – 1 from both sides of the equation (Gujrati, 2004). The testable form of the
Eq. (5.1) is given in Eqs (5.2) and (5.3).
Yt – Yt – 1 = aYt – 1 – Yt – 1 + ut (5.2)
= (a – 1) Yt – 1 + ut
Equation (5.2) can be alternatively written as
DYt = lYt – 1 + ut (5.3)
where, l = (a – 1) and D is the first difference operator.
Therefore, in practice, Eq. (5.3) instead of Eq. (5.2) is estimated, which in
turn, facilitates testing of the null hypothesis whether l = 0 against the alternate
hypothesis l < 0. By doing this, we indirectly test if a = 1, i.e. the series contains
a unit root or not. In other words, whether the series under consideration is
non-stationary or not. In the unit root test, the statistical test used to find if the
estimated coefficient of Yt – 1 (i.e. l) differs significantly from zero, is known
as the Tau (t) statistics or Tau test. In the literature, the Tau statistics or Tau
test is formally known as the Dickey–Fuller (DF) test.
Empirically, the DF test is estimated in different forms considering the
nature of the time series data being analysed. While estimating each one of
the testable forms of DF test, it is assumed that the error term ut is serially
uncorrelated. To resolve the problem of observed serial correlation, if any,
in the error term, Dickey and Fuller (1979) have developed a test known as
100 ● Derivative Markets in India

Augmented Dickey Fuller (ADF) test. In ADF test, the basic DF test equation
is augmented by adding lagged values of the dependent variable DYt . The
ADF test, in our case, will be based on the estimation of Eq. (5.4).
m
DYt = b1 + b 2 t + lYt – 1 + a i ∑ DY
i=1
t–i + et (5.4)

where et is a pure white noise error term, DYt – 1 = (Yt – 1 − Yt – 2), and so on. The
number of lagged variables to be added is determined empirically such that
the error term becomes serially uncorrelated. In ADF test, we still test if l = 0
against the alternate hypothesis, i.e. l < 0.

5.2.2 Testing the Rational Expectations Hypothesis (REH)

In this section, following Stein (1989), we have derived a restriction on average
expected volatilities implied by the options with short-dated and long-dated
maturities, respectively. To derive the restriction, it has been assumed that
the value of the underlying asset (stock price or index value) and volatility
thereof are stochastic processes and the volatility is mean reverting in nature.
The value of the underlying asset and its volatility can be characterized by the
following continuous-time stochastic processes:
dSt = mSt dt + st St dz1 (5.5)
dst2 = – a (st2 – s2) dt + bst dz2 (5.6)
where St is the stock price or index level at time t, m is the mean return on the
stock price or index , st is the volatility of returns on stock/index, dz1 and dz2
are the wiener processes and s 2 is the mean-reversion level1 or long-run average
volatility towards which the instantaneous volatility2 is expected to revert back
(geometrically) and converge in the long-run.
From the Eqs (5.5) and (5.6), the average expected volatility at time t for the
time period T, s 2AV (t, T), can be defined as given in the Eq. (5.7):
1
s 2AV (t, T) = s 2 + (s 2 – s 2 ) [1 – eaT ] (5.7)
aT t
Equation (5.7) clearly explains the mean-reversion process. It indicates that
when the instantaneous volatility st2 is above its mean level volatility s 2 , the
average expected volatility should be decreasing with time to maturity and
vice versa. It may be noted that the instantaneous volatility in the Eq. (5.7) is
not direcltly observable. However, a relationship between average volatilities
differing in maturities only, can be derived by eliminating instantaneous
volatility from both volatility processes (Stein, 1989). Following the argument,
a relationship between average expected volatilities of two differing maturities/
time periods (T1 and T2, where T2 > T1) has been derived. It is generally referred
to as the term structure of average expected volatility, in case volatility is mean
reverting (Stein, 1989). The relationship is given in the Eq. (5.8).
 Testing the Expectations Hypothesis on the Term Structure of Volatilities Implied by........... ● 101

2 T1 ( rT2 – 1)
[s AV (t , T ) – s 2 ] = T1
2
[s AV (t , T1 ) – s 2 ] (5.8)
T2 ( r – 1)
Where s 2AV (t, T1) and s 2AV (t, T2) are the average expected volatility with
short (T1 days) and long (T2 days) maturities respectively, and r is the first
order (daily) autocorrelation coefficient of volatility.
Assuming a constant difference between the maturities (T2 − T1 = DT) of
average expected volatilities, Eq. (5.8) can be expressed in an empirically
testable form (Diz and Finucane, 1993). It is given in Eq. ( 5.9).
2
[s AV 2
(t , T2 ) – s 2 ] = b (T1 , r ) [s AV (t , T1 ) – s 2 ] (5.9)

T1 ( rT2 – 1)
where, b (T1, r) = (5.10)
T2 ( rT1 – 1)
The relationship of the average expected volatilities given in the Eqs (5.9)
and (5.10) can be extended to the implied volatilities estimated from the option
price quotations of at-the-money options including near-the-money3 (NTM)
options. This can be done because the average expected volatility generally
turns out to be approximately equal to the volatility implied by inverting
the Black–Scholes (BS) model for at-the-money options and NTM options
as shown by Hull and White (1987) and Feinstein (1989). Following this, the
relationship of average expected volatilities with differing maturities can be
extended for implied volatilities as well. This relationship has been tested by
Stein (1989), Diz and Finucane (1993), Heynen et al. (1994), Campa and Chang
(1995), Takezaba and Shiraishi (1998) and Poteshman (2001).
Besides, Mixon (2007) has tested the expectations hypothesis on the term
structure of implied volatility of index options for five indices using over-
the-counter options data using Heston’s model. In another study, Das and
Sundaram (1999) made an effort to explain observed shapes of the term
structure of implied volatilities by examining two different volatility models
(jump-diffusion and stochastic volatility).
In case of implied volatilities, the above relationship is arrived at by
2 2
substituting s AV (t , T1 ) and s AV (t , T2 ) by IVtn (the volatility of the underlying
asset implied by short-dated options) and IVtd (the volatility of the underlying
asset implied by long-dated options), respectively. The option price quotations,
for which implied volatilities have been studied as a relationship, are similar
in all respects except the maturity time.
In order to test the restriction empirically, Eq. (5.8) has been examined
by regressing IVtd on the corresponding IVtn (with the same specifications
in terms of strike price, spot price, deal date, etc.) instead of regressing
( IVtd – s –2 ) on ( IVtn – s –2 ) .Equation (5.8) can be estimated in the way
mentioned above because of the ‘free from shift of origin’ property of traditional
102 ● Derivative Markets in India

regression equation, i.e. if we subtract a constant value from both dependent

and independent variables, e.g. s –2 in our case, the results of the estimated
regression equation will remain intact.
Since the property ‘free from shift of origin’ holds only for traditional regression
equation4, which includes an intercept term in the equation, the next challenge
is how to make it applicable in estimation of Eq. (5.8), which is essentially a
regression through origin5, i.e. a model without an intercept term. This can be
done if we could estimate the traditional regression equation in such a way
that the intercept term turns out to be statistically insignificant6, as mentioned
by Gujrati (2004). This can be achieved by taking first difference of both
dependent and independent variables.
The equation that has been estimated empirically is given in the Eq. (5.11).
IVtd = a + b IVtn + e (5.11)
In the above equation, a is the intercept term, b is the empirical elasticity
coefficient7, and e is a white noise error term8. In order to test the REH on the
implied volatilities, b (beta) that connotes the empirical value of the elasticity
coefficient needs to be compared whith the theoretical value of b (beta) derived
on the basis of Eq. (5.10). In Eq. (5.10), amongst all the inputs required for the
calculation of the theoretical value of the beta, r (first order autocorrelation
coefficient of the volatility) is the only unobservable variable and needs a proxy
variable. The proxy variable for the above purpose has been estimated by
calculating the first-order autocorrelation coefficient of the series of volatility
implied by short-dated option contracts, as suggested by Stein (1989).

5.3 DATA AND ESTIMATION OF IMPLIED VOLATILITY

Y

5.3.1 Data
The data considered for the analysis can be broadly classified into three
categories: data related to option contracts; data related to the underlying
asset, i.e. the main index of National Stock Exchange (NSE), formally known
as S&P CNX NIFTY index, and data on risk-free rate of return. The data on
the options consist of daily closing prices of options, strike prices, deal dates,
maturity dates and number of contracts of call and put options. The second
data set, i.e. regarding the underlying asset, includes daily closing value of
S&P CNX NIFTY index and dividend yield on the index. The third data set
consists of monthly average yield on 91-days Treasury-bills. The data for all
the three mentioned categories have been collected from June 04, 2001, (starting
date for index options in Indian securities market) to June 30, 2007.
In order to minimize the bias associated with nonsynchonous trading9, only
liquid option quotations10, i.e. contracts which are having at least 100 contract
traded, are being considered for the analysis. In addition to this, another
 Testing the Expectations Hypothesis on the Term Structure of Volatilities Implied by........... ● 103

filter has also been used. That is, only near-the-money (NTM) contracts have
been selected. In this context, NTM contracts are those contracts that satisfy
a specified range of moneyness (0.90 £ (St/X) £ 1.10) for all the three levels of
time to expiration or maturity, namely (a) time to expiration £ 30 days, (b)
time to expiration 31–60 days and (c) time to expiration 61–90 days.
Besides, another problem related to the presence of multiple contracts
(having same deal date, maturity time and strike price) on a single date. This
required objective selection criteria to zero down such multiple contracts to a
single contract. To address this problem, a short-listing criterion was developed
to select the option quotation that was nearest to the money11 amongst all such
multiple contracts available for a particular date.
To test the REH, two data sets have been prepared across all the level of
moneyness, namely out-of-the-money, at-the-money and in-the-money option
quotations. Each data set for given level of moneyness consists of a pair of
implied volatility (estimated from the option quotations with short-dated and
medium/long-dated maturities12, respectively, which were same in all other
respects, viz. deal date, strike price, etc.). The pair for the first data set (for
a given level of moneyness) consists of (a) implied volatility estimated from
option quotations with time to expiration £ 30 days and (b) implied volatility
estimated from option quotations with time to expiration 31–60 days. Likewise,
another pair of the second data set (for the given level of moneyness) includes
(a) implied volatility estimated from option quotations with time to expiration
£ 30 days and (b) implied volatility estimated from option quotations with time
to expiration 61–90 days. While creating the pairs for the analysis, problem of
non-simultaneity of data have been addressed by selecting only matching data
points. Matching has been done on the basis of date and strike prices.
In addition to this, several short-listing criteria were used to exclude
uninformative option price quotations. These short-listing criteria are:
(i) Options quotations with moneyness outside the range {0.90 £ (St/X)
£ 1.10} in case of a call option and {0.90 £ (X/St) £ 1.10} in case of put
options have been excluded from the analysis. This filtering has been
done because most of the trading volume is concentrated in the options
belonging to the specified range only.
(ii) Secondly, the option quotations which remained untraded (for
which number of contacts traded was zero) have been excluded from
the analysis because these quotations do not contain any relevant
information about the market.
(iii) Thirdly, we have excluded those option quotations for which the lower
boundary condition for option prices has been violated. The violation
of the lower boundary condition occurs when (St − e–r(T – t) X) in case of
call options and (e–r(T – t) X − St) in case of put options, are greater than
the price of call and put options, respectively.
104 ● Derivative Markets in India

5.3.2 Estimation of Implied Volatility

In addition to the model-free approach, select model-based approach has been
proposed to assess the options market efficiency. This has been operationalized
by formulating two objectives addressing options market efficiency using such
approaches. The model-based approach dwells upon some model of options
pricing, e.g. BS model, to have an estimate of implied volatilities, which are
treated as the proxy for all the information traded in the options market.
For the purpose, the volatilities implied by the options prices (implied
volatilities) have been calculated using the adjusted version of BS model, also
known as Black-Scholes-Merton model (1973). The formulae, given in the
Eqs (5.12) and (5.13), incorporate constant dividend yield on the underlying
asset to value the options contract and has been used to estimate the implied
volatilities.
–dT
c = S0 ¥ e N (d1 ) – X ¥ e – rrT N (d2 ) (5.12)
– rT –d T
p= X¥e N ( – d2 ) – S0 ¥ e N ( – d1 ) (5.13)

where,
ln(S0 ¥ e –d T / X ) + (r f + s 2 / 2)T
d1 = ; d2 = d1 – σT1/2,
s T
c and p are price of the call and the put option, respectively, T is the maturity
period, s is the volatility of the underlying asset, S0 is the spot price of the
underlying asset, X is the Strike price of the option contract, d is the constant
continuous dividend yield, rf is the risk-free rate of return and N(x) is the
Normal cumulative probability distribution.
As the BS model has been used to estimate the implied volatilities, it might
give rise to a problem of systematic bias in the estimates of implied volatility
because of its inherent assumptions. It assumes volatility to be constant over
time. On the contrary, time varying behaviour of volatility is well documented
in literature. In addition, the ARCH models (used in analysis) are based on
the empirical premise of time varying volatility. Therefore, the use of BS
model might result into systematic bias in the estimates of implied volatilities.
Fortunately, the problem of systematic bias in the estimates of implied volatility
is less likely to occur as the data on option price quotations that have been
used essentially pertain to the NTM options. Following Claessen and Mittnik
(2002), NTM options have been defined as the options that are up to 10%
in-out-of-the-money. Also, Hull and White (1987) demonstrated empirically
that BS model performs equally well in case of at-the-money options, as the
prices of such option quotations are almost linear in volatilities across all the
maturities.
 Testing the Expectations Hypothesis on the Term Structure of Volatilities Implied by........... ● 105

In this respect, Feinstein (1989) and Takezaba and Shiraishi (1998) made
a similar observation that the implied volatility estimated from BS model
can be used as an unbiased approximation of the expected average volatility
for at-the-money options. Takezaba and Shiraishi (1998) further carried out
the analysis by extending it to the NTM options. In addition, Day and Lewis
(1988 and 1992) also concluded, considering both Black–Scholes (1973) and
Hull and White (1987) model, that the specification error can be minimized
by focusing on at-the-money options because Hull–White and BS models are
linear in average volatility for such options. In view of the above, the present
study has used NTM options in order to minimize the noise in estimating
implied volatilities using BS model.
Besides, efforts have been made to reduce the bias due to non-synchronous
trading, which is likely to distort the quality of implied volatility estimates
derived from the options prices. For the purpose, following Day and Lewis
(1992), all the contracts having at least 100 contracts traded per day have been
chosen for the analysis. The bias due to non-synchronous trading emerges
from the fact that closing prices for the thinly traded contracts are more likely
to represent the transactions that occur before the close of the trading (Day
and Lewis, 1988).
On the selection of appropriate model to estimate the implied volatility,
Stein (1989) mentioned that the use of incorrect model to estimate the implied
volatilities might result into a systematic bias in the estimates. As the model
discussed above has been used to estimate the implied volatilities, it might
give rise to a problem of systematic bias in the estimates of implied volatility
because of its inherent assumptions. The BS model assumes volatility to be a
strictly deterministic function of time. On the contrary, to establish a theoretical
relationship between near-the-month (short-dated) and far-the-month (long-
dated) options, the volatility is assumed to be a stochastic variable. Thus, use
of the BS model might result in systematic bias in the estimates of implied
volatilities and might lessen the authenticity of the results derived from such
estimates of implied volatility.
Fortunately, the problem of systematic bias in the estimates of implied
volatility due the use of the BS model might not occur in our case since the
data on option price quotations that have been used essentially pertain to the
NTM options. Hull and White (1987) in their study have shown empirically
that BS model performs equally well since the prices of such option quotations
are almost linear in volatilities across all the maturities.
Commenting upon this, Feinstein (1989) and Takezaba and Shiraishi (1998)
made a similar observation that the implied volatility, estimated from BS
model, can be used as an unbiased approximation of the expected average
volatility for at-the-money options. Takezaba and Shiraishi (1998) further
carried out the analysis by extending it to the NTM options. This study is
similar to their work since NTM options (with the same definition) have been
the focus of analysis in this chapter. Day and Lewis (1988, 1992) also concluded,
106 ● Derivative Markets in India

considering both Black–Scholes (1973) and Hull and White (1987) model, that
the specification error can be minimized by focusing on at-the-money options
because Hull–White and BS models are linear in average volatility for such
options.
Besides, efforts have been made to reduce the bias due to non-synchronous
trading. The bias due to non-synchronous trading or non-simultaneity has been
offered as a potential explanation of apparent mispricing of option contracts
by Easton (1994). The bias due to non-synchronous trading emerges from
the fact that closing prices for the thinly traded contracts are more likely to
represent the transactions that occur before the close of the trading (Day and
Lewis, 1988).
In the study, the NTM option price quotations refer to those for which the
moneyness indicators, namely (Spot price/Strike price) for the call options
and (Strike price/Spot price) for the put options, ranges from 0.90 to 1.10. The
explanation for the action taken above is quite explicit from the nature of the
data on option price itself, which is presented graphically in Figs 5.1 and 5.2
for call options and put options, respectively.

Figure 5.1 Moneyness and liquidity for Figure 5.2 Moneyness and liquidity for
call option put option

As evidenced by Day and Lewis (1988 and 1992), at NYSE, trading volume
tends to be concentrated in the options that are NTM. It is also apparent
from Figs 5.1 and 5.2 for the Indian options market that 92% and 88% of the
total traded volume pertains essentially to the near-the-month call and put
option price quotations, respectively. Since only near-the-month option price
quotations are being considered for the study, it automatically ensures that
the bias due to non-synchronous trading is minimized. This happens because
any lack of synchronization between the closing index level and closing option
price is less likely to occur.
Following Xu and Taylor (1995), the implied volatilities have been calculated
separately for both call and put options. Besides, following Stein (1989) and
Diz and Finucane (1993), the average implied volatilities (taking an average
of implied volatilities of the underlying estimated from call and put options)
have also been calculated. The reason behind the analysis of average implied
 Testing the Expectations Hypothesis on the Term Structure of Volatilities Implied by........... ● 107

volatility has been the possibility of stale index levels that might induce
potential biases in the estimation of implied volatility when done separately
for call and put options (Diz and Finucane, 1993).
In short, the study focuses on the estimation procedure that assigns greatest
weight on the option price quotations that are affected least by specification
error or bias. This happens because the trading tends to concentrate in NTM
options. This approach is similar to Stein (1989), Harvey and Whaley (1991),
Day and Lewis (1988 and 1992), Takezaba and Shiraishi (1998), Diz and
Finucane (1993) and Campa and Chang (1995).

5.4 ANALYSIS AND EMPIRICAL RESULTS

5.4.1 Descriptive Statistics and Mean-Reversion Property of
implied volatilities
This section deals with the analysis part and some meaningful results drawn
from the data on implied volatilities estimated from index options discussed in
Section 5.3.1. To analyse the data, the methodology described in Section 5.2.1
has been used with a view to test the mean-reversion in implied volatilities;
albeit some other statistical measures have also been used to describe the basic
characteristics of the implied volatility.
The descriptive statistics of implied volatilities have been summarized in
Table 5.1, and Figs. 5.3–5.5 depict the mean volatilities for all the three levels
of maturity pertaining to call options, put options and on the average basis,
respectively. The results indicate that the mean implied volatilities do not
reflect any consistent pattern, as generally warranted by term structure of
implied volatilities, for all the three levels of moneyness and maturity. The
analysis and results so derived can be summarized as:
(a) starting with the out-of-the-money options, the mean implied volatility
in case of call options is increasing with maturity. However, the mean
implied volatilities, when analysed for put options as well as on
average basis (average of implied volatilities estimated from call and
put options), are decreasing first and then increasing again with the
maturity;
(b) In case of at-the-money options, similar to the out-of-the-money options,
the mean implied volatilities are increasing with maturity for the call
options. However, these seem to be decreasing first and then increasing
again with the maturity for put options as well as on average basis;
(c) lastly, in case of in-the-money options, contrary to the (a) and (b), the
mean implied volatilities slightly increase first and then decrease slightly
with the maturity for call options. However, in case of the put options
as well as on average basis, the mean implied volatility is showing a
decreasing trend.
 Table 5.1 Descriptive Statistics for Implied Volatilities of Call, Put and at an Average Level 108
Options
●

Implied Out-of-the-money At-the-money In-the-money

Descriptives
volatility T ≤ 30 31 ≤ T ≤ 60 61 ≤ T ≤ 90 T ≤ 30 31 ≤ T ≤ 60 61 ≤ T ≤ 90 T ≤ 30 31 ≤ T ≤ 60 61 ≤ T ≤ 90
Days Days Days Days Days Days Days Days Days
Mean 0.0485 0.0592 0.0632 0.0460 0.0502 0.0557 0.0525 0.0543 0.0530
Implied Std. Dev. 0.0489 0.1345 0.0706 0.0549 0.0881 0.0452 0.0779 0.0969 0.0482
volatility Skewness 6.81 7.78 4.79 7.45 8.55 2.03 9.74 6.98 4.31
estimated Kurtosis 87.78 68.39 47.58 80.97 88.43 9.53 144.66 58.39 39.73
Derivative Markets in India

from call Jarque–Bera 375377.4 189217.9 40203.4 278533.3 211290.4 654.7604 1029120 125916.1 27279.06
options Probability 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Observations 1222 1005 464 1061 668 266 1208 926 460
Mean 0.0694 0.0678 0.0713 0.0641 0.0586 0.0645 0.0823 0.0675 0.0573
Implied Std. Dev. 0.0841 0.0567 0.0297 0.0675 0.0416 0.0262 0.3996 0.0687 0.0290
volatility Skewness 8.74 7.55 1.26 6.87 3.00 1.79 32.92 9.72 0.65
estimated Kurtosis 117.60 117.49 7.34 80.68 18.10 14.28 1132.11 162.37 5.65
from put Jarque–Bera 735736.70 505128.50 357.86 298411.90 5918.93 484.63 66253583.00 697089.40 42.47
options Probability 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00
Observations 1314 909 341 1151 538 83 1243 649 117
Mean 0.0585 0.0490 0.0529 0.0515 0.0452 0.0542 0.0581 0.0524 0.0499
Std. Dev. 0.0772 0.0313 0.0174 0.0561 0.0239 0.0178 0.0751 0.0284 0.0208
Average Skewness 6.25 2.69 -0.59 6.79 1.66 1.61 6.78 1.58 1.66
implied Kurtosis 56.93 15.64 3.39 81.53 8.24 9.65 66.70 7.03 9.01
volatility* Jarque–Bera 75335.45 2888.26 3.85 133664.10 506.28 138.95 106940.20 610.95 228.00
Probability 0.00 0.00 0.15 0.00 0.00 0.00 0.00 0.00 0.00
Observations 590 367 60 505 315 61 605 558 116
 Testing the Expectations Hypothesis on the Term Structure of Volatilities Implied by........... ● 109

Likewise, the mean-reversion in the implied volatilities of call options, put

options and on average basis has also been examined. To have some-early
evidences of mean-reversion, i.e. whether the series under consideration is
stationary or not, the correlograms (ACFs and PACFs) for all the three levels
of maturity and the moneyness have been plotted separately for call options,
put options, and on their average.

0.075
0.07 Mean IV_Call
0.065 options

0.06 Mean IV_Put

0.055 options
0.05
Mean IV_Average
0.045 basis
0.04
> = 30 Days 31–60 Days 61–90 Days

Figure 5.3 Mean implied volatility estimated from out-of-the-money options

0.07
0.065 Mean IV_Call
options
0.06
0.055 Mean IV_Put
options
0.05
Mean IV_Average
0.045
basis
0.04
> = 30 Days 31–60 Days 61–90 Days

Figure 5.4 Mean implied volatility estimated from at-the-money options

0.09
Mean IV_Call
0.08 options
0.07
Mean IV_Put
options
0.06
Mean IV_Average
0.05 basis

0.04
>=30 Days 31–60 Days 61–90 Days

Figure 5.5 Mean implied volatility estimated from in-the-money options

110 ● Derivative Markets in India

In addition to correlograms, ADF test has been applied for all the three levels
of maturity and the moneyness separately for call options, put options and on
their average. The results of ADF test are reported in Tables 5.2, 5.3 and 5.4 for
call options, put options and on their average basis, respectively. In order to
make error term a pure white noise, the number of lags of dependent variable
to be added in the equation of ADF test has been decided on the basis of Akaike
information criterion (AIC) and/ or Schwarz information criterion (SIC).
The results of the ADF test clearly demonstrate that the implied volatilities
do not contain a unit root. The null hypotheses that the series of implied
volatilities are having a unit root have been tested at 5% as well as 1% level
of significance. In the tables, the significance levels provided in parentheses
(p-values) along with the empirical value of tau statistic connote that almost
all the implied volatilities for call options, put options and on their average are
significant at even 1% level of significance except a few, which are significant
only at 5%. The rejection of the null hypotheses that the implied volatilities
are having a unit root makes the series of implied volatilities stationary, i.e.
the implied volatilities, in fact, are mean reverting in nature.
The stationarity so observed validates mean-reversion in implied volatilities
and indicates that the investors price options considering mean-reversion
property of the volatility. However, the mean reversion in itself does not
ensure that the expectations regarding the evolution of volatilities are being
formed in the way as warranted by REH. In view of this, the empirical test
of the restriction has been carried out too, and the results are summarized in
the next section.

5.4.2 Test of the Rational Expectations Hypothesis (REH) on

Implied Volatilities
In this section, the REH is tested on the basis of methodology discussed in
Section 5.2.2. Test of the RHE can be classified into three stages, namely (a)
estimations of empirical elasticity coefficients by regressing implied volatilities
estimated from the long-dated option on the corresponding implied volatilities
estimated from the short-dated options for all possible pairs discussed in
Section 5.3.1; (b) calculation of theoretical elasticity coefficient for the same
and (c) comparison of empirical elasticity coefficient against the theoretical
elasticity coefficient.
The estimation of empirical elasticity coefficient involves regression of
the implied volatility estimated from long-dated option on corresponding
implied volatility estimated from the short-dated options. We have estimated
the elasticity coefficient for all the possible pairs discussed in the Section
5.3.1. Besides, another classification (within pair) that has been made for
the regression analysis is on the basis of the constant difference between the
maturities of long-dated and corresponding short-dated option quotations.
 Table 5.2 Testing Mean-Reversion (Stationarity) of the Implied Volatilities of Call Options
Augmented Dickey Fuller t/t-test
S.
Description of options Empirical value Critical value of t Inferences
no.
of t at a = 5% at a = 1% at a = 5% at a = 1%
− 6.1577
1 Near-the-month call option − 2.8637 − 3.4355 Mean reverting Mean reverting
(0.0000)
− 7.2353
2 Far-the-month1 call option − 2.8642 − 3.4367 Mean reverting Mean reverting
(0.0000)
− 3.0198

Out-of-money
3 Far-the-month2 call option − 2.8676 − 3.4444 Mean reverting Non-mean reverting
(0.0338)
− 5.9855
4 Near-the-month call option − 2.8641 − 3.4363 Mean reverting Mean reverting
(0.0000)
− 8.2594
5 Far-the-month1 call option − 2.8657 − 3.4399 Mean reverting Mean reverting
(0.0000)
− 3.2777

At-the-money
6 Far-the-month2 call option − 2.8724 − 3.4553 Mean reverting Non-mean reverting
(0.0169)

[0.90 £ (S0/X) £ 1.10]

− 8.4945

Near-the-money options
7 Near-the-month call option − 2.8637 − 3.4356 Mean reverting Mean reverting
(0.0000)
− 30.4295
8 Far-the-month1 call option − 2.8644 − 3.4372 Mean reverting Mean reverting
(0.0000)
− 21.4476

In-the-money
9 Far-the-month2 call option − 2.8676 − 3.4443 Mean reverting Mean reverting
(0.0000)
12
Note: The figures in parentheses show ‘level of significance’ / ‘p value’.

1
Options having 31–60 days to maturity.
2
Options having 61–90 days to maturity.
Testing the Expectations Hypothesis on the Term Structure of Volatilities Implied by...........
●
111
 Table 5.3 Testing Mean-Reversion (Stationarity) in the Implied Volatilities of Put Options 1 2 112
Augmented Dickey Fuller t/t-test
●

S.
Description of options Empirical value Critical value of t Inferences
no.
of t at a = 5% at a = 1% at a = 5% at a = 1%
− 6.1981
1 Near-the-month put option − 2.8638 − 3.4358 Mean reverting Mean reverting
(0.0000)
− 3.9069
2 Far-the-month1 put option − 2.867 − 3.4423 Mean reverting Mean reverting
(0.0021)
− 6.4729

Out-of-money
3 Far-the-month2 put option − 2.8972 − 3.5123 Mean reverting Mean reverting
Derivative Markets in India

(0.0000)
− 35.4276
4 Near-the-month put option − 2.8636 − 3.4354 Mean reverting Mean reverting
(0.0000)
− 24.4757
5 Far-the-month1 put option − 2.8658 − 3.4402 Mean reverting Mean reverting
(0.0000)
− 5.6345

At-the-money
6 Far-the-month2 put option − 2.8658 − 3.4876 Mean reverting Mean reverting
(0.0000)

[0.90 ≤ (S0/X) ≤ 1.10]

− 5.6886

Near-the-money options
7 Near-the-month put option − 2.8635 − 3.4351 Mean reverting Mean reverting
(0.0000)
− 4.5421
8 Far-the-month1 put option − 2.8645 − 3.4373 Mean reverting Mean reverting
(0.0002)
− 2.8796

In-the-money
9 Far-the-month2 put option − 2.8699 − 3.4496 Mean reverting Non-mean reverting
(0.0488)
1
Options having 31–60 days to maturity.
2
Options having 61–90 days to maturity.
 Table 5.4 Testing Mean-Reversion (Stationarity) of the Average Implied Volatilities*
Augmented Dickey Fuller t/t-test
S.
Description of options Empirical value Critical value of t Inferences
no.
of t at a = 5% at a = 1% at a = 5% at a = 1%
−8.6483
1 Near-the-month option −2.8662 −3.4412 Mean reverting Mean reverting
(0.0000)
−18.3613
2 Far-the-month1 option −2.8692 −3.4480 Mean reverting Mean reverting
(0.0000)
−3.7731

Out-of-money
3 Far-the-month2 option −2.9117 −3.5461 Mean reverting Mean reverting
(0.0053)
−4.7362
4 Near-the-month option −2.8671 −3.4432 Mean reverting Mean reverting
(0.0001)
−16.2688
5 Far-the-month1 option −2.8705 −3.4510 Mean reverting Mean reverting
(0.0000)
−8.7377

At-the-money
6 Far-the-month2 option −2.9109 −3.5441 Mean reverting Mean reverting
(0.0000)

[0.90 ≤ (S0/X) ≤ 1.10]

−14.3506

Near-the-money options
7 Near-the-month option −2.8661 −3.4410 Mean reverting Mean reverting
(0.0000)
−24.9652
8 Far-the-month1 option −2.8665 −3.4419 Mean reverting Mean reverting
(0.0000)
−11.9199

In-the-money
9 Far-the-month2 option −2.8867 −3.4881 Mean reverting Mean reverting
(0.0000)
*
Average implied volatility = (Implied volatility of call option + Implied volatility of put option)/2 1 2
1
Options having 31–60 days to maturity.
2
Options having 61–90 days to maturity.
Testing the Expectations Hypothesis on the Term Structure of Volatilities Implied by...........
●
113
114 ● Derivative Markets in India

This has been done because the calculation of theoretical elasticity coefficients
requires that the difference of maturities has to be constant (Diz and Finucane,
1993). The classification of implied volatilities so arrived at as per levels of
moneyness, maturities and constant maturity differences is presented in Tables
5.5a, 5.5b, 5.6a, 5.6b, 5.7a and 5.7b, respectively.
While estimating the empirical elasticity coefficients, the traditional
regression equation (including an intercept term) has been estimated in such a
way that the intercept term does not come out to be significantly (statistically)
different from zero, or in other words, the intercept term is zero, and the
residual or error term becomes a white noise. To do the same, the methodology
that has been adopted is ARIMA (n, 1, 0)13, i.e. a traditional regression equation,
which includes n lags of dependent variable as independent variables as well
as first order differencing in order to get a zero intercept term.
The precondition to estimate the ARIMA model is that the series under
consideration should be stationary. Since the prerequisite of stationarity of
implied volatilities have already been ensured in the Section 5.4.1, the next
issue in the estimation of ARIMA model is to decide upon the number of
lags to be included in the model. To respond to the issue, ACF and PACF
for the implied volatilities have been plotted to have an initial idea about the
probable lag structure of the model. Besides, AIC and SIC have been used to
select the best model with a zero intercept term and a white noise error term.
In addition to this, the calculation of coefficient of determination has been done
separately for the regression pairs of average basis implied volatility. This has
been done to draw a comparison with the similar study done by Stein (1989).
The coefficient of determination has been calculated as15:
Residual sum of squares
R2 = 1 –
Total sum of squares
After estimating the ARIMA (n, 1, 0), we have also examined the error
term to ensure that the model has fitted the data well, and the error term is a
white noise. In order to ensure that the error term is a white noise, it has been
examined whether (a) the error term is serially uncorrelated up to the one forth
of its lag length and (b) its mean is zero. To test the serial independence in
error term, the correlograms and a statistics put forth by Ljung and Box (1978)
have been used. To test the mean value of the error term, t-test was used to
see whether it is statistically different from zero. The estimates of empirical
elasticity coefficients so arrived are given in Tables 5.5b, 5.6b and 5.7b.
Next to the estimation of the empirical elasticity coefficients, calculation
of the theoretical coefficients of elasticity was done. The calculation of the
theoretical elasticity coefficients has been done on the basis of the formula given
in the Eq. (5.10). From the formula, it is clear that all the variables required
for the calculation of the theoretical coefficients of the volatilities are known
except one variable—the autocorrelation coefficient of the volatilities.
 Table 5.5 (a) Calculation of the Theoretical Elasticity Coefficient of the Implied Volatilities of Call Options
Constant maturity Average maturity of Average maturity of First order auto Theoretical elasticity
Options difference3 short-dated options (T1) long-dated options (T2) correlation coefficient coefficient
(in days) (in days) (in days) (r) (b)
CALL OTM 28 14.61 42.61 0.766 0.350
(30–60) 35 10.15 45.16 0.663 0.228
CALL OTM 56 16.78 72.78 0.601 0.231
(30–90) 63 13.11 76.16 0.55 0.172
CALL ATM 28 14.65 42.65 0.716 0.346
(30–60) 35 16.33 45.35 0.291 0.360
CALL ATM 56 18.06 74.06 0.531 0.244
(30–90) 63 12.5 75.64 0.251 0.165
CALL ITM 28 14.31 42.32 0.618 0.338
(30–60) 35 11.3 46.32 0.024 0.244
CALL ITM 56 17 73 0.529 0.233
(30–90) 63 11.86 74.92 0.093 0.158
3
Constant maturity difference signifies the difference between maturities of ‘near-the-month’ and ‘far-the-month’ options. Such differences have been calculated
for each pair, i.e., between ‘8–30 days to maturity and 31–60 days to maturity’, and between ‘8–30 days to maturity and 61–90 days to maturity’. Moreover,
within each pair, the data has been arranged in two sub-groups for further analysis.
Table 5.5 (b) Theoretical and Empirical Elasticity Coefficients of the Implied Volatilities of Call Options
Values of elasticity coefficient of mean-reversion (b) under REH
Constant maturity difference*
Options Empirical values from Conclusion
(in days) Theoretical values
detrended series
CALL OTM 28 0.350 No causality NA
(30–60) 35 0.228 0.183 Underreaction*
Testing the Expectations Hypothesis on the Term Structure of Volatilities Implied by...........
●

CALL OTM 56 0.231 2.596 Overreaction*

(30–90) 63 0.172 0.922 Overreaction
115
 Values of elasticity coefficient of mean-reversion (b) under REH 116
Constant maturity difference*
Options Empirical values from Conclusion
●

(in days) Theoretical values

detrended series
CALL ATM 28 0.346 0.496 Overreaction
(30–60) 35 0.360 0.305 Underreaction
CALL ATM 56 0.244 1.796 Overreaction
(30–90) 63 0.165 0.488 Overreaction
CALL ITM 28 0.338 0.656 Overreaction
(30–60) 35 0.244 0.032 Underreaction
Derivative Markets in India

CALL ITM 56 0.233 1.261 Overreaction

(30–90) 63 0.158 0.320 Overreaction
*When the long-dated volatility turns out to be more/less than that is expected based on short-dated volatility, assuming rational expectations to hold, this is
termed as Overreaction/ Underreaction of long-dated volatility.

Table 5.6 (a) Calculation of the Theoretical Elasticity Coefficient of the Implied Volatilities of Put Options
Constant maturity Average maturity of Average maturity of First order auto Theoretical elasticity
Options difference* short-dated options (T1) long-dated options (T2) correlation coefficient coefficient
(in days) (in days) (in days) (r) (b)
PUT OTM 28 13.01 41.01 0.774 0.329
(30–60) 35 9.99 45.02 0.285 0.222
PUT OTM 56 18.16 74.16 0.697 0.245
(30–90) 63 13.49 76.6 0.39 0.176
PUT ATM 28 12.85 40.84 0.683 0.317
(30–60) 35 9.15 44.17 0.631 0.210
PUT ATM 56 18.72 74.72 0.415 0.251
(30–90) 63 11.57 74.57 0.565 0.155
PUT ITM 28 12.63 40.62 0.547 0.311
(30–60) 35 9.52 44.53 0.699 0.221
 Constant maturity Average maturity of Average maturity of First order auto Theoretical elasticity
Options difference* short-dated options (T1) long-dated options (T2) correlation coefficient coefficient
(in days) (in days) (in days) (r) (b)
PUT ITM
63 14.25 75.89 0.365 0.188
(30–90)

Table 5.6 (b) Theoretical and Empirical Elasticity Coefficients of the Implied Volatilities of Put Options
Values of elasticity coefficient of mean-reversion (b) under REH
Constant maturity difference
Options Empirical values from Conclusion
(in days) Theoretical values
detrended series
PUT OTM 28 0.329 0.433 Overreaction
(30–60) 35 0.222 0.186 Underreaction
PUT OTM 56 0.245 1.221 Overreaction
(30–90) 63 0.176 0.238 Overreaction
PUT ATM 28 0.317 0.497 Overreaction
(30–60) 35 0.210 0.289 Overreaction
PUT ATM 56 0.251 0.565 Overreaction
(30–90) 63 0.155 0.559 Overreaction
PUT ITM 28 0.311 0.412 Overreaction
(30–60) 35 0.221 0.230 Overreaction
PUT ITM
63 0.188 0.275 Overreaction
(30–90)

Table 5.7 (a) Calculation of the Theoretical Elasticity Coefficient of the Average Implied Volatilities
Constant maturity Average maturity of Average maturity of First order auto Theoretical elasticity
Options difference* short-dated options (T1) long-dated options (T2) correlation coefficient coefficient
Testing the Expectations Hypothesis on the Term Structure of Volatilities Implied by...........

(in days) (in days) (in days) (r) (b)

●

OTM 28 12.88 40.86 0.753 0.324

(30–60) 35 9.3 44.32 0.404 0.210
117
 Constant maturity Average maturity of Average maturity of First order auto Theoretical elasticity 118
Options difference* short-dated options (T1) long-dated options (T2) correlation coefficient coefficient
●

(in days) (in days) (in days) (r) (b)

OTM
56 14.69 73.3 0.252 0.200
(30–90)
ATM 28 11.62 39.65 0.564 0.293
(30–60) 35 8.43 43.47 0.44 0.194
ATM
56 18.52 72.74 0.528 0.255
(30–90)
ITM 28 13.25 41.27 0.633 0.322
Derivative Markets in India

(30–60) 35 10.81 45.82 0.236

0.529
ITM
63 16.84 73.31 0.462 0.230
(30–90)

Table 5.7 (b) Theoretical and Empirical Elasticity Coefficients of Average Implied Volatilities
Constant maturity Values of elasticity coefficient of mean-reversion (b) under REH
Options difference* Empirical values from R2 (%) Conclusion
(in days) Theoretical values
detrended series
OTM 28 0.324 0.801 88.67 Overreaction
(30–60) 35 0.210 0.388 85.29 Overreaction
OTM
56 0.200 0.252 72.60 Overreaction
(30–90)
ATM 28 0.293 0.667 85.51 Overreaction
(30–60) 35 0.194 0.450 66.92 Overreaction
ATM
56 0.255 0.528 73.68 Overreaction
(30–90)
ITM 28 0.322 0.296 79.59 Underreaction
(30–60) 35 0.236 0.372 63.72 Overreaction
ITM
63 0.230 0.462 72.00 Overreaction
(30–90)
 Testing the Expectations Hypothesis on the Term Structure of Volatilities Implied by........... ● 119

As suggested by Stein (1989), the estimated autocorrelation coefficient of

implied volatilities estimated from the short-dated options can be used as
a proxy for the autocorrelation coefficient of average volatilities. Thus, the
autocorrelation coefficient estimated from the short-dated implied volatilities
was used as a proxy for autocorrelation coefficient in the calculation of the
theoretical elasticity coefficient.
Though use of a proxy autocorrelation coefficient might lessen the
authenticity of the results, relatively less sensitivity of theoretical coefficient to
the change in the autocorrelation coefficient (as explicit from the formula given
in Eq. 5.10) reduces the possibility of such outcomes. As tested empirically,
even large variations in the value of the estimated autocorrelation coefficient
are no how affecting the decisions regarding the rational expectations. That is,
the theoretical elasticity coefficient will not be significantly affected by even
large variations in the value of autocorrelation coefficient, as evident from Eq.
5.10. This further motivated us to accept estimated autocorrelation coefficient
of implied volatilities as a proxy for autocorrelation coefficient of short-dated
average volatility. The calculation of the theoretical elasticity coefficients is
given in the Tables 5.5a, 5.6a and 5.7a.
After the estimation of the empirical and theoretical elasticity coefficients,
a comparison has been drawn. The comparison drawn is given in the Tables
5.5b, 5.5b and 5.7b. The results so arrived clearly indicate that the implied
volatility of long-dated options, in general, is overstated vis-à-vis the expected
volatility of long-dated options, estimated on the basis of implied volatility
of corresponding short-dated options assuming rational expectations to hold.
As explicitly given in the tables, the implied volatility of long-dated options
is overreacting, since 8 out of 11 pairs in case of implied volatility of call
option, 10 out of 11 in case of implied volatility of put options and overall 8
out of 9 in case of the average implied volatility seem to be biased in favour of
overreaction. The findings are in line with the findings of the study done by
Stein (1989) but just opposite to the results of Poterba and Summers (1986).
The overreaction of long-dated implied volatility seems to be more
pronounced because in majority of the cases, the empirical coefficients are
more than double compared whit theoretical coefficients and in few cases it is
even as high as 10 times of the theoretical elasticity coefficient. Apart from this,
the autocorrelation coefficients of short-dated implied volatility are relatively
small compared with those in the study done by Stein (1989).
The observation clearly differentiates the behaviour of Indian investors
from their counterparts in the developed world. From the findings, it can be
observed that the Indian investors seem to give less importance to the volatility
of short-dated options for forecasting the volatilities of corresponding long-
dated options to value the option contracts since the R2 values are coming out
to be relatively less compared with the study done by Stein (1989).
120 ● Derivative Markets in India

Besides, in most of the cases, values of empirical elasticity coefficients

seem to be much lower than unity and, thus, in contrast with the findings of
Stein (1989), where these came out to be quite close to unity in majority of the
cases. These observations reinforce that Indian investors are not assigning
much importance to the volatility of short-dated options while arriving at an
estimate of long-dated volatility to value a long-dated option contract unlike
their counterparts in the developed world.

5.5 CONCLUDING OBSERVATIONS

The study reveals that the implied volatility estimated from the short-dated
as well as long-dated options is mean reverting. This is a good sign for the
development of derivatives market in India. However, the mean implied
volatility does not exhibit a consistent pattern across all the three categories;
namely call options, put options and their average basis. The patterns of
volatility are similar for put and options based on the average basis. The mean
implied volatility in case of call options has shown an opposite pattern.
Another notable finding is that the implied volatilities estimated from
the long-dated index options do not evolve as expected (assuming rational
expectations to hold) on the basis of volatility implied by the short-dated index
options. The results, in general, depict severe violation of REH on the term
structure of implied volatilities. These findings are in conformity with those
of Stein (1989) and Byoun et al. (2003). In specific terms, the violation of the
hypothesis has been seen in terms of the overreaction of long-dated implied
volatility; the conclusion is quite similar to that of Stein (1989).
In addition to this, relatively weak coefficients of determination and
empirical autocorrelation coefficients, compared with that in Stein (1989),
indicate that Indian investors assign relatively less importance to the volatility
implied by short-dated option to arrive at an estimate of the volatility to be
used in valuing long-dated options. In operational terms, it indicates that the
Indian investors either do not use term structure approach to price the option
contracts, or they might not be aware of this due to the inadequate experience
of derivatives market.
In a nutshell, it may be reasonable to conclude that the Indian investors are
not exhibiting rational behaviour while valuing index options. This indicates
price inefficiency in index options market in India and might have serious
impact on the development of derivatives market, in general, and options
market, in particular.

END NOTES
1. Mean-reversion level connotes that level of long-run average volatility to
which the instantaneous volatility is expected to return in a long period
of time.
Testing the Expectations Hypothesis on the Term Structure of Volatilities Implied by........... ● 121

2. This refers to the current level of volatility at any specified point of time.
3. In the study, near-the-money (NTM) options stand for those option
contracts that lie within the specified range (0.90 ≤ (S0/X) ≤ 1.10) of
moneyness.
4. A regression equation which includes an intercept term and regression
coefficients along with the corresponding independent variables to
explain the variations in the dependent variable.
5. A regression equation which does not include an intercept term or a
traditional regression equation where the intercept terms comes out to
be zero/statistically insignificant.
6. An acceptable percentage of error while rejecting the null hypothesis
that is expected to occur purely due to chance. The most preferable
levels of significance are 5% and 1%.
7. Empirical elasticity coefficient denotes the degree of causal relationship
between the short-dated and long-dated volatility that is observed
empirically by regressing long-dated volatility on the corresponding
short-dated volatility.
8. An error term is said to be a white noise when it is normally distributed
and remains serially uncorrelated up to a significant lag length.
9. In this context, non-synchronous trading refers to the phenomenon of
different closing timings of the two markets, i.e. the options market and
the underlying’s market.
10. In the study, the liquid options stand for those option contracts, which,
at least, have one contract traded.
11. Nearest-to-the-money options imply those option quotations which,
within the specified range of moneyness, are most close to one (at-the-
money option) or, in other words, the option quotation for which the
difference between their strike price and spot price is the least amongst
all given on a particular date and level of maturity. For example, if on
July 15, 2007, when the spot price stands at ` 4257.5; there are three
contracts with the strike prices ` 4250, 4255 and 4265, respectively
(assuming that all the contracts are having same maturity). The contract
with the strike price ` 4255 will be said to be nearest-to-the-money.
12. In the study short-dated maturity refers to the period of less than equal
to the time period of 30 days, a time period between 31days to 60 days
belongs to the medium-dated option contracts, and Long-dated option
contracts cover the time period of greater than 60 days and less than or
equal to 90 days.
13. ARIMA (n, 1, 0) stands for Autoregressive Integrated Moving average
model with n autoregressive terms in the model as independent
variables, with first order level of differencing.
122 ● Derivative Markets in India

14. In the formula, residual sum of squares is given in the SPSS generated
solution for ARIMA models in the table called residuals diagnostics
and total sum of squares has been calculated by measuring variation in
the long-dated implied volatility (dependent variable) around its mean
value.
*Average implied volatility = (Implied volatility of call option + Implied volatility
of put option)/2
6 C H A P T E R

Informational Efficiency of
Implied Volatilities of S&P
CNX Nifty Index Options

6.1 INTRODUCTION
The implied volatilities (IVs) of options contracts represent the market’s ex-ante
forecast of the average volatility of the underlying asset over the remaining
life of the option contract (Merton, 1973; Hull and White, 1987). The IVs have
been of equal interest to academics as well as practitioners because of the
information they contain about the near future of the market and their usage
for valuation of options, risk hedging, portfolio selection, etc. These aspects
facilitate in assessing the informational efficiency of the options market in that
whether the options market is performing satisfactorily on its well-identified
function, viz. risk hedging, price discovery in the underlying’s market and
allocation of capital, as pointed out by Ackert and Tian (2001).
The informational efficiency of IVs has been tested by a number of studies
across the globe. For this purpose, it is hypothesized that the IVs impound
all the information contained in the historical returns. Therefore, in an
informational efficient options market, the forecasts based on IVs should
outperform the forecasts based on the historical returns. In view of this,
majority of the studies have compared IVs as a future estimate of volatility with
forecasts based on select conditional volatility models, namely Generalized
Autoregressive Conditional Heteroscedasticity (GARCH) and Exponential
GARCH or EGARCH models.
In literature, some of the studies have supported the hypothesis that IVs
are informational efficient, i.e. these can be used as a predictor of future
volatility, whereas some of the studies have rejected this hypothesis. In this
respect, Day and Lewis (1992), a pioneering study on informational efficiency
of IVs of Standard & Poor’s 100 index options market of USA, found that the
IVs could not impound all the information available in the past returns and,
therefore, concluded that the IVs were not better forecasts of futures volatility
compared with those based on select conditional volatility models, namely,
GARCH(1, 1) and EGARCH(1, 1). The inability of implied volatility to forecast
124 ● Derivative Markets in India

future volatility is indicative of informational inefficiency of options market.

Lamoureux and Lastrapes (1993), who investigated this issue using stochastic
volatility model, for the very first time, on several individual stocks traded
on Chicago Board Options Exchange (CBOE), and Canina and Figlewski
(1993) also rejected the IVs as a predictor of future volatility in S&P 100 index
options market.
In contrast, a number of empirical studies found support for the hypothesis
that the IVs can be used as a predictor for future volatility and historical returns
do not contain any information beyond the IVs (Latane and Rendleman, 1976;
Chiras and Manaster, 1978; Gemmill, 1986; Shastri and Tandon, 1986; Scott
and Tucker, 1989). Also, some recent studies on the informational efficiency
of options market, e.g. Blair et al. (2001) and Claessen and Mittnik (2002),
supported the hypothesis and concluded that IVs impound all the information
available in past returns data. In addition, Xu and Taylor (1995) and Guo (1996)
found support for the hypothesis in PHLX currency options market.
In this study, the informational efficiency of IVs, calculated from daily
closing premium of S&P CNX Nifty index options, is tested vis-à-vis the
forecasts from select conditional volatility models, namely GARCH(1,1) and
EGARCH(1,1). For the purpose, the methodology proposed by Day and Lewis
(1992) constitutes the basis for the study. In the study, in-the-sample as well
as out-of-the-sample forecast abilities of IVs have been tested by regressing
volatility forecasts (estimated using the select conditional volatility models
on historical returns of S&P CNX Nifty index) on IVs. In addition to this,
the study attempts at a comparative analysis of informational efficiency of
IVs of call and put options unlike majority of studies, which have focused
on the call options only, and draws its relationship with the present market
microstructure in India.
The rest of the chapter is organized in five sections. Characteristics of
volatility and conditional volatility models have been discussed in Section 6.2.
Section 6.3 contains data and calculation of implied volatility. Theoretical
framework for examining informational efficiency of IVs has been discussed
in Section 6.4. Analysis and results are presented in Section 6.5. The chapter
ends with concluding observation in Section 6.6.

6.2 CHARACTERISTICS OF VOLATILITY AND CONDITIONAL

VOLATILITY MODELS

6.2.1 Characteristics of Volatility

In literature, a number of features of financial time series and stock market
volatility have been identified across the globe. Some of the select features
of volatility of time series data that have been found commonly across the
 Informational Efficiency of Implied Volatilities of S&P CNX Nifty Index Options ● 125

world are volatility clustering, mean-reversion, fat-tailed or leptokurtic distribution

of returns and the leverage effect.
1. Volatility clustering: Volatility clustering refers to the tendency of large
changes in asset prices (of either sign) to follow large changes and small
changes (of either sign) to follow small changes (Brooks, 2005. In another
study, Mandelbrot (1963) noted that the periods of large returns were
clustered and distinct from periods of small returns, which were also
clustered. This indicates that the current level of volatility is positively
correlated with that of the immediately preceding period(s).
Therefore, it would be appropriate to infer that volatility changes over
time (i.e. heteroscedastic in nature) and exhibits clusters of large and
small changes. Also, this characteristic has been reported by numerous
other studies, e.g. Fama (1965), Chou (1988), Schwert (1989) and Baillie
et al. (1996). In short, it signifies that the volatility today is expected to
influence the level of volatility in future.

Figure 6.1 Volatility clustering in daily returns of S&P CNX Nifty index, June 1997–2006

For clustering, data consisting of NSE Nifty index values has been
analysed for the period of 10 years from June 2, 1997 to June 30, 2006.
Figure 6.1 shows the daily returns of S&P CNX Nifty index and the
typical pattern of clustering. The returns are expressed in percentage
terms and are continuously compounded [ln(It/It – 1); where It and
It – 1 represent the value of index at day t and the previous day, t – 1,
respectively]. The figure clearly indicates that there are different phases
of high as well as low volatility. In the present study, this tendency in
index returns has been confirmed by using Ljung-Box statistics.
2. Fat-tailed distribution of asset returns: The fat-tailed distribution
property of asset returns is also known as leptokurtic (highly peaked)
distribution of returns. This represents the tendency of returns to cluster
126 ● Derivative Markets in India

excessively near the mean value (i.e. more peaked around the mean)
and higher probability of returns at tails of the distribution compared
with expected returns in case of a mesokurtic (normal) distribution. In
earlier studies, it is documented in Mandelbrot (1963) and Fama (1965).
Nattenberg (1994) concludes that stock returns exhibit non-normal
skewness and kurtosis. These findings were also supported by the work
of Corrado and Su (1997), Clark (1973) and Blattberg and Gonedes (1974).
In this respect, Ghysels et al. (1996) contend that volatility clustering
and fat tails of asset returns are intimately related. The distributional
assumption of returns/innovations becomes important as it needs to be
specified for the estimation of variance equation in conditional volatility
models, e.g. GARCH and EGARCH.
3. Mean reversion: Mean reversion in volatilities indicates the tendency of
volatility to return to its normal level (i.e. the long-run level of volatility)
in long-run. It signifies that irrespective of the magnitude and sign of
fluctuations, the series reverts to its mean. Such a series in time-series
statistics is refereed to as a covariance stationary series, i.e. a series
which has a constant mean, variance and autocovariance (at different
lags), regardless of the point of time at which the measurement is made.
Though the property of mean reversion is accepted by most of the
practitioners, they may differ in the magnitude of the normal volatility
level and changes in it over a period of time (Engle and Patton, 2001).
The mean reversion in IVs has been examined empirically by Dixit et
al. (2007), a study on Indian options market.
4. Leverage effect: It is designated as the main reason for the asymmetrical
contribution of innovations (having different sign) to the volatility
of returns of a financial time series. In financial markets, it has been
observed that a negative shock contributes more to the volatility of
returns compared with a positive shock. That is, the volatility is high
when the market falls and is found to be relatively lower in times
of upward trends. This is explained by leverage effect. The financial
leverage or debt-equity ratio of the company increases as the value of
equity shares goes down in the market. This causes increase in the risk
the investor’s perceive as the futures streams of the cash flows from
the company become more uncertain. Therefore, a down movement
in the stock prices is expected to cause more volatility compared with
appositive movement. This effect was first noted by Black (1976) and
later supported by Christie (1982), Schwert (1989), Nelson (1991), Glosten
et al. (1993) and Engle and Ng (1993). They observe that the changes in
stock prices tend to be negatively correlated with the changes in stock
return volatility, and changes in stock return volatility are too large in
response to the changes in returns direction.
 Informational Efficiency of Implied Volatilities of S&P CNX Nifty Index Options ● 127

These typical characteristics of the financial time series data have helped
financial econometricians to engineer more sophisticated models for
measurement and forecasting of volatility and led to the conditional volatility
models for more precise measurement and estimation of volatility.

6.2.2 Conditional Volatility Models

6.2.2.1 GARCH model
Generalized ARCH (GARCH) models of volatility, introduced by Bollerslev
(1986), came into existence in order to overcome the problems of ARCH model
of Engle (1982). This model estimates variance as the weighted average of past
squared residuals but has declining weights that never go completely to zero
(Engle, 2002). GARCH is a parsimonious version of the ARCH model, which
is achieved by adding moving average term(s) or GARCH component(s) and,
essentially, can be written as an ARCH model with infinite lag structure. In this
respect, Brooks (2005) opined that these models are more parsimonious than
ARCH model and avoid over-fitting. In addition, non-negativity constraints1
are less likely to be violated, as GARCH provides a parsimonious model.
The GARCH(1, 1) models with AR(1) mean equation is given in Eqs (6.1) and
(6.2).
rt = c + frt – 1 + et et ~ N(0, st2)  [Mean equation] (6.1)
st2 = w + a e 2t – 1 + b s 2t – 1) [Variance equation] (6.2)
where,
c is a constant term, f is the first-order autoregression coefficient, et is the
normally distributed error term/innovations with zero mean and time varying
(heteroscedastic) variance. w = l * VL, where VL is the long-run/unconditional
variance of the series and l = 1 – a – b . st2 and s 2t – 1 are the GARCH estimates
of variance for the period t and t − 1, respectively. a and b are the ARCH and
GARCH parameters, respectively.
This model represents variance for the current period as a weighted average
of three sources of variance, namely (a) long-run or unconditional variance
(VL); (b) new information about volatility or innovations during the previous
period (e 2t – 1), i.e. the ARCH component; and (c) the conditional variance for the
last period (s 2t – 1) or the GARCH component estimated at point of time t − 2.
The most widely used GARCH specification, i.e. GARCH(1,1), asserts that
the best predictor of the variance in next period is a weighted average of the
long-run average variance, the variance predicted for this period and the
new information in this period that is captured by the most recent residual
(Engle, 2002). This model can be extended to a GARCH(p, q) specification,
where current conditional variance is parameterized to depend upon q lags
of the squared error and p lags of the conditional variance. The GARCH(p, q)
model is summarized in Eq. (6.3.)
128 ● Derivative Markets in India

q p
s 2t = w+ ∑
i =1
a i e t2– i + ∑b s
j =1
j
2
t–j (6.3)

Diagnostics of GARCH models

In order to ascertain the effectiveness of the specified model, a number of
diagnostic tests have been specified. To test the mean equation, one can test the
series ofet et, Engle (2002). The test of mean equation attempts to test whether
{et} is a white noise or not, i.e. whether it has a constant mean and variance.
For this purpose, in general, correlograms, i.e. Autocorrelation Functions
(ACFs) and Partial Autocorrelation Functions (PACFs) along with Ljung-Box
(Q) test, are used up to a specified lag length to find how successfully the
specified mean equation has modelled the data. In general, 15 lags can be used
to diagnose the problems in specification, if any (Engle, 2002). For a correctly
specified mean equation, the ACFs and PACFs along with the Ljung-Box (Q)
statistics should turn out to be insignificant up to the specified lags. Similarly,
squared innovations {et2} are used to test the correct specification of the variance
equation in the model. In addition, ARCH LM test up to a specified lag—1 for
a GARCH(1,1) variance equation—is carried out to diagnose the remaining
ARCH effect in the data that could not be captured/explained by the model.
For a correctly specified variance equation, the ACFs, PACFs, Q statistics
along with the ARCH LM test should turn out to be insignificant (up to the
specified lag length).

Limitations of GARCH models

In spite of numerous attractive qualities that the GARCH models possess,
these are subject to certain limitations as well. Harvey (1981) reported a few
limitations of GARCH models. Some of these limitations are: (a) GARCH
models do not account for leverage effect as the conditional variance
incorporates the same impact of innovations (et) irrespective of the sign they
possess, and (b) these models might result in violation of non-negativity
constraints on parameters.

6.2.2.2 EGARCH model

In order to overcome the above-mentioned limitations, Nelson (1991)
introduced EGARCH model, where conditional variance is constrained to be
non-negative by specifying the logarithm of st2 to be a function of the past
es(et–1, et –2 ..., et – n ). In EGARCH specification, the conditional variance depends
upon the magnitude as well as sign of the lagged residual. The conditional
variance equation for the EGARCH(1,1) model is given in Eq. (6.4).
et – 1 et – 1
In s 2t = w + b ln(s t2– 1 ) + a +g (6.4)
st – 1 st – 1
 Informational Efficiency of Implied Volatilities of S&P CNX Nifty Index Options ● 129

In the Eq, (6.4), coefficient g represents the leverage effect. Empirically, g

should possess a negative sign and be statistically significant to corroborate
the leverage effect of innovations on volatility. From Eq. (6.4), it can be inferred
that a positive et–1 contributes less to the volatility compared with a negative
et–1 as g is expected to possess a negative sign if the leverage effect is present
in the series being analysed. Notably, this specification enables st2 to respond
asymmetrically to rises and falls in the error term and has several advantages
over GARCH model, viz. (a) st2 will always be positive even if the parameters
are negative, as the log specification is used in this model, thus avoiding the
need to impose the non-negativity constraints, and (b) this model accounts
for the asymmetries as γ becomes negative in case the relationship between
volatility and return is negative.
Besides, there are other models that respond to the asymmetrical
contribution of innovations to the variance equation. These include Threshold
GARCH (TGARCH) introduced by Zakoian (1991), Quadratic GARCH
(QGARCH) proposed by Sentana (1995), Semi-parametric ARCH model of
Engle and Gonzalez-Rivera (1991), and Log GARCH models of Pantula (1986)
and Geweke (1986), among others. In addition, Engle and Ng (1993) gave the
concept of News Impact Curve that depicts the impact of new information on the
next period variance. The standard GARCH model has this curve symmetrical;
it implies that positive and negative surprises of the same magnitude would
produce the same amount of volatility. However, a negative innovation
causes more volatility than a positive innovation of the same size, and thus,
a GARCH model results in under-prediction of volatility following bad news
and over-prediction following good news.

6.3 DATA AND IMPLIED VOLATILITY

6.3.1 Data
The data considered for the analysis can consist of (i) data related to S&P CNX
Nifty index options contracts, (ii) data related to the underlying asset, i.e. the
S&P CNX Nifty index, and (iii) data on the risk-free rate of return. The data
on the options consist of daily closing prices of options, strike prices, deal
dates, maturity dates and number of contracts of call as well as put options. In
order to minimize the bias associated with nonsynchonous trading, only liquid
option quotations (i.e. contracts which are having at least 100 contracts traded)
have been considered for the analysis. The second data set is regarding the
underlying asset. It includes daily closing prices of S&P CNX Nifty index. The
third data set consists of monthly average yield on 91-days Treasury-bills with
maturity date most close to the expiry of the options contracts.
The details pertaining to data collection and conversion of discrete yields
into continuous one have been provided in Chapter 3.
130 ● Derivative Markets in India

6.3.2 Implied Volatilities

The volatilities implied by the options prices (IVs) have been calculated using
the adjusted version of Black-Scholes (BS) model, also known as Black-Scholes-
Merton model (1973). The Black-Scholes formula and other details on IVs have
been provided in Chapter 3.
As the BS model has been used to estimate the IVs, it might give rise to a
problem of systematic bias in the estimates of implied volatility because of its
inherent assumptions. It assumes volatility to be constant over time. On the
contrary, time varying behaviour of volatility is well documented in literature.
In addition, the ARCH models (used in analysis) are based on the empirical
premise of time varying volatility. Therefore, per-se, the use of BS model might
result into systematic bias in the estimates of IVs. Fortunately, the problem of
systematic bias in the estimates of implied volatility is less likely to occur, as
the data on option price quotations that have been used essentially pertain
to the near-the-money options. And, Hull and White (1987) demonstrated
empirically that BS model performs equally well in the case of at-the-money
options as the prices of such option quotations are almost linear in volatilities
across all the maturities.
Feinstein (1989) and Takezaba and Shiraishi (1998) made a similar
observation that the implied volatility, estimated from BS model, can be
used as an unbiased approximation of the expected average volatility for
at-the-money options. Takezaba and Shiraishi (1998) further carried out the
analysis by extending it to the near-the-money options. In addition, Day and
Lewis (1988, 1992) also concluded, considering both Black–Scholes (1973) and
Hull and White (1987) model, that the specification error can be minimized
by focusing on at-the-money options because Hull–White and BS models are
linear in average volatility for such options. In view of the above, the present
study has used near-the-money options in order to minimize the noise in
estimating IVs using BS model. Following Claessen and Mittnik (2002), near-
the-money options have been defined as the options that are up to 10% in/
out of-the-money.
Moreover, the shortest maturity options, having maturity time of 8–30 days,
have been chosen in order to reduce the possibility of maturity mismatch. The
options having minimum 8 days to maturity have been chosen in view of the
fact that the possibility of the week-end effect on IVs could be ruled out. The
maturity period chosen is similar to various studies conducted in this area,
e.g. Xu and Taylor (1995) and Claessen and Mittnik (2002).
Besides, efforts have been made to reduce the bias due to non-synchronous
trading, which is likely to distort the quality of implied volatility estimates
derived from the options prices. For this purpose, following Day and Lewis
(1992), all the contracts having at least 100 contracts traded per day have been
 Informational Efficiency of Implied Volatilities of S&P CNX Nifty Index Options ● 131

chosen for the analysis. The bias due to non-synchronous trading emerges
from the fact that closing prices for the thinly traded contracts are more likely
to represent transactions that occur before the close of the trading (Day and
Lewis, 1988). For further details refer to Chapter 5.

6.4 INFORMATIONAL EFFICIENCY OF IMPLIED VOLATILITY

VIS-À-VIS VOLATILITIES ESTIMATED USING ARCH VOLATILITY
MODELS
In the study, an attempt has been made to test the informational efficiency of
IVs of call as well as put options. The IVs are said to be informational efficient
when no other model of volatility estimation is able to capture any information
(based on the historical returns) beyond the informational contents of IVs. In
other words, the IVs shall contain all the information available in historical
returns if the options market is informational efficient.
To this end, the informational efficiency of IVs has been tested vis-à-vis the
two popular conditional volatility models, namely GARCH and EGARCH.
The conditional volatility models have been chosen as the forecasts from such
models have less variance compared with those from unconditional models
(Enders, 2005. Moreover, the ARCH models become a natural choice for the
purpose because (i) these are able to capture much of the volatility clustering
and serial correlation, well documented in the literature on financial time
series, and (ii) the estimation of ARCH model within the maximum likelihood
framework is straightforward. In this respect, commenting upon the functional
utility of ARCH models amongst other studies in the literature, Busch (2005)
opines that the GARCH model has become an extremely popular tool for
modelling conditional variance amongst academics and practitioners.
In view of the above, the efficiency tests of the IVs have been conducted
based on the two models, namely GARCH(1,1)-IV and EGARCH(1,1)-IV. That
is to say, the scope of the present study is confined to testing the IVs vis-à-vis
GARCH(1,1) and EGARCH(1,1) models only.

6.4.1 GARCH(1,1)-IV and EGARCH(1,1)-IV Models

The GARCH(1,1)-IV model attempts to test the informational efficiency of
IVs vis-à-vis the GARCH(1,1) model. The model incorporates daily implied
volatility/variance from the options contracts as an additional explanatory
variable in the variance equation. The proposed model, i.e. AR(1)-GARCH(1,1)-
IV, which has been used for in-the-sample analysis, is summarized in Eqs (6.5)
and (6.6). Though the model mentioned in Eqs (6.5) and (6.6) is based on the
literature reviewed, it has been validated empirically for the data used in the
present study for in-the-sample as well as out-of-the-sample analysis.
rt = c + frt–1 + et; et ~ GED (0, st)2 (6.5)
132 ● Derivative Markets in India

s 2t = w + a e 2t–1 + b s 2t–1 + dIV2t–1 (6.6)

In Eqs (6.5) and (6.6), all the elements are the same as in Eqs (6.1) and
(6.2) except the distributional assumption of innovations/error terms, i.e.
generalized error distribution (GED). Notably, Eq. (6.6) includes another
exogenous variable, i.e. implied volatility, in addition to those included in
Eq. (6.2). In other words, the model given in Eq. (6.2) represents a constrained
version of the model given in Eq. (6.6), with the constraint d = 0.
As discussed earlier, for an options market to be informational efficient, all
the information available in historical returns should be reflected in IVs, or in
this case, forecasts from the GARCH model should not reflect or contain any
information beyond the IVs. Therefore, in operational terms, the ARCH and
GARCH parameters (a and b, respectively) should become zero in presence of
implied volatility as an exogenous variable in the GARCH variance equation,
if the options market has to be informational efficient. In short, in the GARCH
(1, 1)-IV model, the null hypothesis tested is that the ARCH and GARCH
coefficients are zero (H0: a and b = 0).
In addition, the EGARCH model has been used to test whether the IVs
incorporate the leverage effect or not. In other words, the EGARCH-IV model
has been proposed to test if the IVs respond asymmetrically to the innovations
with different sign (+ve and –ve) or not. The informational efficiency
(particularly the presence of leverage effect) of IVs vis-à-vis volatility estimates
using EGARCH models has been analysed for in-the-sample data using the
model summarized in Eq. (6.6). For the model, the innovations (et) have been
derived from the model for mean equation, summarized in Eq. (6.5).
et – 1 et – 1
ln (s 2t ) = w + b ln(s t2– 1 ) + a +g + d ln( IVt2– 1 ) (6.7)
st – 1 st – 1
Equation (6.7) can be seen as an extension of Eq. (6.4) by adding implied
volatility as an additional exogenous variable. In the model, it has been
hypothesized that all the three parameters a, b and g = 0, if the options market
is informational efficient. The EGARCH-IV model tests the null hypothesis that
g = 0, which is additional to the GARCH-IV model (a and b = 0). This hypothesis
addresses to the leverage effect, i.e. the IVs incorporate the asymmetrical effect
of negative and positive shocks in the returns of the underlying asset.
The methodology adopted, i.e. the introduction of implied volatility as an
exogenous variable in the GARCH and EGARCH variance equation, is in line
with a number of studies carried out across the globe, including the pioneering
one by Day and Levis (1992) on S&P 100 index options in the context of the
US market. Some other studies which adopted similar methodology include
Lamoureux and Lastrapes (1993), Canina and Figlewski (1993), Xu and Taylor
(1995), Fleming (1998), Claessen and Mittnik (2002), Pong et al. (2004).
 Informational Efficiency of Implied Volatilities of S&P CNX Nifty Index Options ● 133

Though the above-mentioned methodology has been used extensively

for assessing the informational efficiency of IVs, one major problem of this
methodology has been the maturity mismatch as documented by some studies,
e.g. Xu and Taylor (1995), Claessen and Mittinik (2002).
The problem of maturity mismatch emanates from the fact that the two
forecasts to be compared, viz. the forecasts from ARCH models and implied
volatility based forecasts, represent different maturities. The forecast from the
IVs represent average forecast of volatility for the rest of the life of the options,
whereas the ARCH models generate forecasts for the next period (one day,
in our case, for in-the-sample analysis and one-week for out-of-the-sample
analysis). Xu and Taylor (1995), in their study on currency options market,
empirically examined the effect of maturity mismatch on the performance of
mixed models (ARCH-IV models). They concluded that the maturity of implied
volatility does not affect the performance of the ARCH-IV models.

6.5 ANALYSIS AND EMPIRICAL RESULTS

6.5.1 Behavior of Stock Returns and ARCH Effect

The estimation of an ARCH model requires three basic specifications—(i) the
order of the mean equation, (ii) distributional assumption for innovations and
(iii) specification of the ARCH variance equation. In this section, an attempt
has been made to diagnose the correct specification for the innovations to be
used in the ARCH variance equation. For the purpose, all the six years’ data
(i.e. from June 04, 2001 to June 30, 2007) have been analysed to arrive at the
correct distributional assumption for the innovations.
Table 6.1 Summary Statistics of the Daily Returns of S&P CNX Nifty Index and
the Diagnosis for ARCH Effect, June 04, 2001, to June 30, 2006
Summary statistics of returns
No. of Std. Jarque–
Mean Skewness Kurtosis Probability
observations Dev. Bera
0.0490
2772 1.5975 −0.3892 7.7817 2710.82 0.000
(0.1060)
Ljung–Box Statistics ARCH LM test
Lag
rt |rt | rt2 TR2
4 20.29 (0.000) 344.40 (0.000) 381.03 (0.000) 317.17 (0.000)
8 29.36 (0.000) 549.79 (0.000) 449.02 (0.000) 336.82 (0.000)
12 48.44 (0.000) 666.43 (0.000) 483.76 (0.000) 342.41 (0.000)
16 56.13 (0.000) 741.11 (0.000) 509.09 (0.000) 345.52 (0.000)
32 79.04 (0.000) 946.59 (0.000) 565.52 (0.000) 364.50 (0.000)
The most commonly used specification for the innovations is the normal
distribution. However, in practice, the GED has been found to be a more correct
134 ● Derivative Markets in India

specification for the financial time serieshaving fat-tails. In order to arrive at the
correct specification for the innovations, the assumption of normal distribution
has been tested empirically, and the results are summarized in Table 6.1.
Skewness and kurtosis of the returns data provide initial evidences that
the distribution of returns departs from normality, as the observed values
of skewness and kustosis are different from the desired values of 0 and 3,
respectively, required for a normal distribution. The departure from normality
further gets corroborated by the Jarque and Bera test (summarized in Table
6.1), as it indicates severe departure from the normality. In addition, the
kurtosis of 7.7817 is indicative of leptokurtic distribution of returns, i.e. high
peakedness of the data and a relatively higher likelihood of observations on
the tails (fat-tails) compared with a normal (mesokurtic) distribution. The
high peakedness and fat-tails for the returns data are clearly depicted in
Fig. 6.2, which compares the empirical distribution of returns with the normal
distribution for such data set.

Figure 6.2 Histogram of returns and normal distribution

These findings clearly indicate that the assumption of normal distribution

for the returns/innovations would not be appropriate. And, some other
distributional assumption should be specified to respond to the behaviour
of the data. For the purpose, it is imperative to assume that the returns/
innovations follow the GED, which is the most suited distribution for the
data having high peakedness and fat-tails. This assumption has further been
validated while estimating the coefficient of variance equation for the ARCH
models.
6.5.1.1 Testing the ARCH effect
Moreover, another very popular characteristic of the financial time series data,
i.e. the presence of ARCH, has been validated using Ljung–Box (Q) statistics
and the Engle’s ARCH-LM test statistics. The results are summarized in Table
6.1. The results of the Q statistics (for the returns, absolute returns and squared
 Informational Efficiency of Implied Volatilities of S&P CNX Nifty Index Options ● 135

returns data series) and ARCH test have been found to be highly significant
at the all specified lags, viz. 4, 8, 12, 16 and 32. The high significance levels
for Q statistics at all specified lags for returns and absolute returns series
indicate the presence of serial correlation in the data. The high significance
of Q statistics for the squared returns, prima-facie, suggests the presence of
conditional heteroscedaticity. The presence of ARCH effect in the data further
gets validated by the high significance levels for Engle’s ARCH-LM test.
In short, the financial time series under consideration demonstrates the
typical characteristics which suggest an ARCH model with GED innovations,
an obvious choice for forecasting the volatility of NSE CNX Nifty index
returns.

6.5.2 In-The-Sample Results

The results regarding in-the-sample analysis of the informational efficiency
of IVs of call as well as put options vis-à-vis the volatility estimates from the
select ARCH models are summarized in Tables 6.2 and 6.3. Table 6.2 contains
the results using GARCH(1,1) model, and Table 6.3 presents results using
EGARCH(1,1) model.

6.5.2.1 ARCH & GARCH effect and informational efficiency of implied

volatility
The results summarized in Table 6.2 reveal that the IVs could not capture all
the information available in GARCH forecasts for call as well as put options.
It is evident from the fact that the GARCH coefficient (b) in GARCH-IV model
turned out to be significant at 5% level of significance for both call and put
options. In addition, notwithstanding the IVs of call options, the IVs of put
options failed to capture the ARCH effect as the coefficient representing ARCH
effect (a) turned out to be statistically significant at 5% level of significance in
the GARCH-IV model for put options. The findings, prima-facie, are indicative
of informational inefficiency of the IVs for both call and put options.
Despites being informational inefficient, IVs seem to have incorporated
some relevant information contained in GARCH forecasts as the persistence
of volatility, i.e. (a + b), has reduced from 0.8357 (in the constrained model, i.e.
GARCH model) to 0.5175 (in the unconstrained model, i.e. GARCH-IV model)
along with the statistically significant coefficient of implied volatility, 0.4354.
Likewise, the persistence of volatility in the case of put options has reduced
from 0.9606 to 0.7526 and implied volatility turned out to be a significant
predictor, however, with relatively lower coefficient, i.e. 0.1107.
Moreover, the incremental informational contents of IVs have been
examined by comparing the two models, i.e. the unconstrained model3 and the
constrained model4. For the purpose, Likelihood Ratio (LR) test has been carried
out. The results of LR test are summarized in Table 6.2. The results clearly
 136
●

Table 6.2 In-The-Sample Test of Informational Efficiency of Implied Volatilities of Call as well as Put Options vis-à-vis Volatility Estimates
Using GARCH(1,1) Model, June 04, 2001, to June 30, 2006 (Unconstrained Model: s 2t = w + ae 2t –1 + bs 2t –1 + dIV2t –1 )
Options Coefficients GED Log Likelihood
Model
type w a b d parameter likelihood Ratio (LR) test
Derivative Markets in India

Unconstrained model 0.000040 0.0762 0.4413 0.4354 1.2160

2644.44
(GARCH-IV model) (0.043) (0.055) (0.049) (0.046) (0.000)
Call 21.86
Constrained model
Options 0.000038 0.1314 0.7045 Constrained 1.1884 (P < 0.01)
(GARCH-IV model with the 2633.51
(.007) (.000) (.000) to zero (0.000)
constraint d = 0)
Unconstrained model 0.000034 0.1318 0.6208 0.1107 1.1676
2353.05
(GARCH-IV model) (0.035) (0.000) (0.000) (0.030) (0.000)
Put 5.02
Constrained model
Options 0.000010 0.0584 0.9022 Constrained 1.1550 (P < 0.05)
(GARCH-IV model with the 2350.54
(0.050) (0.050) (0.050) to zero (0.000)
constraint d = 0)
(i) The figures given in parentheses denote level of significance (p-values).
(ii) To estimate the coefficient of the ARCH variance equation, the conditional mean equation which has been found adequate is
rt = c + f1 rt – 1 + et, where et ~ GED (0, ht).
 Informational Efficiency of Implied Volatilities of S&P CNX Nifty Index Options ● 137

indicate that the IVs do have some relevant information. This is borne out by
the fact that the introduction of IVs as an exogenous variable in the GARCH
variance equation resulted in an increase in the value of log likelihood functions
(LLF), and the increase was found to be statistically significant.
Notably, the IVs of call options seem to be less inefficient compared with
those of put options, as the latter failed to capture ARCH as well as GARCH
effect present in the GARCH forecasts compared with the former which failed
to include only the GARCH effect. This finding further gets corroborated by
the fact that the coefficient of IVs in GARCH-IV models for call options turns
out to be higher compared with that in the case of put options, i.e. 0.4354 and
0.1107 for call and put options, respectively.
Besides, the incremental information of IVs in the case of put options seem
to be relatively less since its introduction as an exogenous variable in the
GARCH variance equation has increased the LLF just by 5.02 compared with
an increase of 21.86 in the case of call options. Also, the increase in LLF has
been found to be significant at 5% level of significance compared with the call
options, where it is significant even at 1% level of significance. In sum, ‘in-
the-sample’ analysis using GARCH model reveals that the IVs of call options
have been informationally superior to those of put options.

6.5.2.2 Leverage effect and informational efficiency of implied volatility

This study has also attempted to diagnose the leverage effect in the IVs, as the
informational efficiency requires that IVs should impound all the information
contained in historical returns. For the purpose, EGARCH-IV model has
been used. The results of EGARCH(1,1) model, summarized in Table 6.3,
clearly demonstrate that Indian securities market exhibits leverage effect in
volatility of returns, as the parameter addressing the leverage effect (g ) turns
out to be statistical significant and has a negative sign attached to it. That is,
the Indian market, in line with its international counterparts, corroborates
that a negative shock adds more to the volatility of returns compared with a
positive shock.
The informational inefficiency of IVs is reaffirmed by the results of the
EGARCH-IV model, as all the coefficients of the EGARCH model (in conformity
with the results using GARCH model) remained statistically significant even
after the inclusion of IVs as an exogenous variable in the EGARCH variance
equation for call as well as put options; however, the ARCH coefficient (a )
was found to be statistically insignificant at 5% level of significance for call
options. The results of the model reject the null hypothesis of informational
efficiency of the IVs, i.e. a , b and g = 0.
The most crucial finding from the EGARCH-IV model is that the IVs have
failed to incorporate the leverage effect as well (in addition to the ARCH and
GARCH effect). This is borne out by the fact that the coefficient g remains
138 ● Derivative Markets in India

negative and statistically significant in the unconstrained model for both call
and put options. These findings, prima-facie, lead to the conclusion that the IVs
are informational inefficient. In sum, it would be reasonable to conclude that
the IVs have failed to incorporate all the typical characteristics of the returns
series under consideration, for in-the-sample analysis.
However, the informational inefficiency of IVs does not necessarily mean
that these do not include any information contained in EGARCH forecasts.
It is evident from the fact that the persistence of volatility for call options has
reduced from 0.8410 to 0.6649 on account of inclusion of implied volatility as
an exogenous variable in the variance equation. A reduction in persistence of
volatility was also observed in the case of put options, as it has reduced from
0.8785 to 0.7841. Besides, the statistical significance of the coefficient associated
with IVs denotes that these do have incremental information. The finding
that the IVs, in spite of their informational inefficiency, contain some useful
information has further been corroborated by the significant values of the LR
tests. Notably, the coefficients of IVs are statistically significant; however, the
magnitude is very low compared with those in the case of GARCH model. This
could be attributed to the inability of IVs to capture the leverage effect.
Moreover, as far as comparative efficiency of IVs of call and put options
is concerned, the results are in line with the GARCH-IV model. The poor
performance of IVs of put options is indicated by comparatively lower
magnitude of coefficient of IVs of put options in EGARCH-IV variance
equation. Also, the increase in LLF on account of introduction of implied
volatility as an exogenous variable is 5.80 for put options compared with an
increase of 15.18 for call options. Therefore, it would be reasonable to conclude
that put options market is more inefficient compared with call options market
in India.

6.5.3 Out-Of-The-Sample Forecast

In addition to ‘in-the-sample’ examination of IVs, it would be appropriate to
look at their ‘out-of-the-sample’ performance to have a complete assessment
of the informational efficiency. For the purpose, the ex-ante forecasting ability
of IVs has been examined vis-à-vis ex-ante forecasts from GARCH(1,1) and
EGARCH(1,1) models. In the study, a sample period of 1 year, i.e. from July
03, 2006 to June 29, 2007, has been used to assess the out-of-sample forecast
ability of all the volatility models. The period of 1 year has been chosen from
the total data of 6 years as the correct estimation of ARCH models requires
a larger sample. Therefore, the data for the first 5 years have been used to
estimate the coefficients of ARCH model(s) required for forecasting ex-ante
volatility for the very first week. Thereafter, a rolling sample for 5 years data
has been used for ex-ante forecast of subsequent weeks. The approach is
similar to Xu and Taylor (1995), which used 5 years data from the total data
 Table 6.3 In-The-Sample Test of Informational Efficiency of Implied Volatilities of Call as well as Put Options vis-à-vis Volatility Estimates Using EGARCH
et – 1 et – 1

+
+
=
(1,1) Model, June 04, 2001 to June 30, 2006 [Unconstrained Model: ln (s t2 ) = w b ln(s t2–1) a +g + d ln(IVt2– 1) ]
st – 1 st – 1

Options Coefficients GED Log Likelihood

Model
type w a g b d parameter likelihood Ratio (LR) test
Unconstrained model −1.4615 0.1277 −0.1878 0.6649 0.1654 1.2239
2648.18
(EGARCH-IV model) (0.002) (0.052) (0.000) (0.000) (0.013) (0.000)
Call 15.18
Options Constrained model
−1.49795 0.1978 − 0.1513 0.8410 Constrained 1.2103 (P < 0.01)
(EGARCH-IV model with 2640.59
(.002) (.001) (.000) (.000) to zero (0.000)
the constraint d = 0)
Unconstrained model −1.2099 0.1899 − 0.1813 0.7841 0.0890 1.1839
2358.64
(EGARCH-IV model) (0.001) (0.000) (0.000) (0.000) (0.013) (0.000)
Put 5.80
Constrained model
Options −1.1809 0.2076 − 0.1265 0.8785 Constrained 1.1810 (P < 0.05)
(EGARCH-IV model with 2355.74
(0.001) (0.000) (0.001) (0.000) to zero (0.000)
the constraint d = 0)
(i) The figures given in parentheses denote level of significance (p-values).
(ii) To estimate the coefficient of the ARCH variance equation, the conditional mean equation which has been found adequate is
rt = c + f1 rt –1 + et, where et ~ GED (0, ht).
Informational Efficiency of Implied Volatilities of S&P CNX Nifty Index Options
●
139
140 ● Derivative Markets in India

of 7 years to estimate the ARCH coefficients for forecasting ex-ante volatility

for the rest of the period.
A total of 53 non-overlapping one-week ahead ex-ante forecasts of volatility
have been examined, as the forecasting horizon chosen is one-week. The
weekly forecasts from ARCH models have been carried out on rolling basis
with 5-years daily data. The rolling data set included most recent daily data
available just before the week for which forecast is to be made. For example,
if the forecast is to be made for the week ‘t’, the data set includes all the most
recent daily observations for a period of 5 years up to the last trading day of
week ‘t − 1’. Similarly, for the forecast for week ‘t + 1’, the data set has been
revised by including all the daily observations for the most recent week, i.e.
‘t’, and the data pertaining to the oldest week has been excluded so that the
sample size remains intact. The coefficients of the 53 weekly revised ARCH
models (based on these rolling data samples) estimated for forecasting the
ex-ante one-week ahead forecast are summarized in Table 6.4.
Table 6.4 Coefficients of ARCH Models Estimated on the Basis of Weekly
Revised Rolling Samples Having 5 Years Daily Data
Coefficients

S. Estimation et – 1 e t –1
s 2t = w + ae 2t–1 + bs 2t–1 ln(s t2 ) = w + a +g
2
+ b ln(s t –1 )
no. date s t –1 s t –1

w a b w a g b
1 30-Jun-06 0.000013* 0.1559* 0.7746* −1.0473* 0.2706* −0.1369* 0.9054*
2 7-Jul-06 0.000012* 0.1530* 0.7780* −1.0333* 0.2707* −0.1340* 0.9071*
3 14-Jul-06 0.000012* 0.1546* 0.7771* −1.0274* 0.2739* −0.1318* 0.9080*
4 21-Jul-06 0.000012* 0.1546* 0.7820* −0.9915* 0.2736* −0.1316* 0.9119*
5 28-Jul-06 0.000012* 0.1548* 0.7811* −0.9911* 0.2707* −0.1330* 0.9117*
6 4-Aug-06 0.000013* 0.1588* 0.7714* −1.0294* 0.2668* −0.1515* 0.9070*
7 11-Aug-06 0.000012* 0.1575* 0.7741* −1.0279* 0.2707* −0.1384* 0.9076*
8 18-Aug-06 0.000013* 0.1581* 0.7719* −1.0359* 0.2685* −0.1419* 0.9065*
9 25-Aug-06 0.000013* 0.1658* 0.7636* −1.0435* 0.2688* −0.1471* 0.9054*
10 1-Sep-06 0.000013* 0.1608* 0.7690* −1.0515* 0.2695* −0.1436* 0.9049*
11 8-Sep-06 0.000013* 0.1594* 0.7719* −1.0777* 0.2642* −0.1556* 0.9013*
12 15-Sep-06 0.000014* 0.1583* 0.7639* −1.1392* 0.2619* −0.1537* 0.8940*
13 22-Sep-06 0.000013* 0.1528* 0.7740* −1.1256* 0.2600* −0.1506* 0.8955*
14 30-Sep-06 0.000012* 0.1429* 0.7896* −1.0417* 0.2543* −0.1381* 0.9048*
15 6-Oct-06 0.000013* 0.1472* 0.7743* −1.0719* 0.2543* −0.1345* 0.9015*
16 13-Oct-06 0.000012* 0.1414* 0.7880* −1.0649* 0.2557* −0.1319* 0.9024*
17 20-Oct-06 0.000012* 0.1400* 0.7912* −1.0582* 0.2545* −0.1337* 0.9031*
18 27-Oct-06 0.000012* 0.1401* 0.7912* −1.0532* 0.2550* −0.1325* 0.9037*
19 3-Nov-06 0.000012* 0.1418* 0.7896* −1.0518* 0.2573* −0.1312* 0.9041*
20 10-Nov-06 0.000012* 0.1442* 0.7872* −1.0635* 0.2601* −0.1348* 0.9031*
 Informational Efficiency of Implied Volatilities of S&P CNX Nifty Index Options ● 141

Coefficients

S. Estimation et – 1 e t –1
s 2t = w + ae 2t–1 + bs 2t–1 ln(s t2 ) = w + a +g
2
+ b ln(s t –1 )
no. date s t –1 s t –1

w a b w a g b
21 17-Nov-06 0.000011* 0.1434* 0.7920* −1.0254* 0.2611* −0.1308* 0.9076*
22 24-Nov-06 0.000011* 0.1469* 0.7860* −1.0857* 0.2683* −0.1320* 0.9014*
23 1-Dec-06 0.000011* 0.1436* 0.7919* −1.0252* 0.2605* −0.1286* 0.9076*
24 8-Dec-06 0.000011* 0.1424* 0.7938* −1.0453* 0.2569* −0.1388* 0.9051*
25 15-Dec-06 0.000012* 0.1506* 0.7818* −1.1328* 0.2631* −0.1515* 0.8956*
26 22-Dec-06 0.000012* 0.1506* 0.7822* −1.1150* 0.2627* −0.1502* 0.8975*
27 29-Dec-06 0.000012* 0.1540* 0.7752* −1.1371* 0.2647* −0.1512* 0.8952*
28 5-Jan-07 0.000012* 0.1538* 0.7760* −1.1685* 0.2716* −0.1486* 0.8924*
29 12-Jan-07 0.000012* 0.1572* 0.7724* −1.1588* 0.2689* −0.1540* 0.8932*
30 19-Jan-07 0.000012* 0.1565* 0.7716* −1.1881* 0.2703* −0.1537* 0.8901*
31 26-Jan-07 0.000012* 0.1570* 0.7707* −1.1824* 0.2699* −0.1546* 0.8907*
32 2-Feb-07 0.000012* 0.1564* 0.7715* −1.1669* 0.2680* −0.1541* 0.8923*
33 9-Feb-07 0.000012* 0.1565* 0.7719* −1.1570* 0.2660* −0.1575* 0.8932*
34 16-Feb-07 0.000013* 0.1555* 0.7714* −1.1941* 0.2640* −0.1638* 0.8887*
35 23-Feb-07 0.000012* 0.1542* 0.7780* −1.1653* 0.2638* −0.1622* 0.8920*
36 2-Mar-07 0.000012* 0.1650* 0.7668* −1.1869* 0.2727* −0.1634* 0.8902*
37 9-Mar-07 0.000012* 0.1658* 0.7694* −1.1409* 0.2688* −0.1640* 0.8950*
38 16-Mar-07 0.000011* 0.1594* 0.7782* −1.1406* 0.2719* −0.1575* 0.8955*
39 23-Mar-07 0.000011* 0.1603* 0.7777* −1.0829* 0.2643* −0.1608* 0.9013*
40 30-Mar-07 0.000011* 0.1597* 0.7770* −1.0881* 0.2634* −0.1612* 0.9006*
41 6-Apr-07 0.000010* 0.1522* 0.7930* −1.0185* 0.2576* −0.1551* 0.9078*
42 13-Apr-07 0.000010* 0.1533* 0.7914* −1.0269* 0.2583* −0.1581* 0.9069*
43 20-Apr-07 0.000011* 0.1583* 0.7848* −1.0724* 0.2680* −0.1563* 0.9027*
44 27-Apr-07 0.000010* 0.1529* 0.7914* −1.0153* 0.2553* −0.1579* 0.9078*
45 4-May-07 0.000010* 0.1519* 0.7931* −1.0151* 0.2564* −0.1546* 0.9080*
46 11-May-07 0.000011* 0.1585* 0.7838* −1.0505* 0.2661* −0.1518* 0.9050*
47 18-May-07 0.000011* 0.1584* 0.7844* −1.0490* 0.2647* −0.1539* 0.9051*
48 25-May-07 0.000011* 0.1590* 0.7834* −1.0473* 0.2638* −0.1551* 0.9051*
49 1-Jun-07 0.000011* 0.1577* 0.7853* −1.0385* 0.2620* −0.1567* 0.9059*
50 8-Jun-07 0.000010* 0.1497* 0.7949* −1.0125* 0.2584* −0.1514* 0.9086*
51 15-Jun-07 0.000010* 0.1494* 0.7929* −1.0514* 0.2623* −0.1456* 0.9046*
52 22-Jun-07 0.000010* 0.1484* 0.7956* −1.0289* 0.2614* −0.1429* 0.9071*
53 29-Jun-07 0.000010* 0.1500* 0.7946* −1.0267* 0.2637* −0.1436* 0.9076*
* represents that the associated coefficient is significant at 1% level of significance.
Note: (1) To estimate the coefficient of the ARCH variance equation, the conditional mean equation
which has been found adequate is r1 = c + f1 rt – 1 + f4 rt – 4 + et, where et ~ GED (0. ht).
142 ● Derivative Markets in India

On the basis of the estimated ARCH coefficients, the ex-ante forecasts of

volatility for a total number of 53 weeks have been determined. The weekly
forecast represents the average volatility for the week. That is, it is the average
of the daily forecasted volatilities for all trading days available in that week.
Equation (6.8) represents the formula for calculating the weekly average
variance.
n
1 D

∑h
W
hF , t = F,t + i (6.8)
n i =1
W
where, h F , t is the average ex-ante forecast of conditional variance for the
D
week at time t; h F , t + 1 is the daily ex-ante forecast of conditional variance for
the ith day of the week at time t estimated from GARCH(1,1) or EGARCH(1,1)
model and n represents the number of trading days in that week. All such
forecasts of volatility have been annualized multiplying by 252 (number of
trading days in the year).
Likewise, the IVs of the options with shortest maturity (ranging between
‘8–30 days to maturity’) have been used as an ex-ante forecast of volatility
(V1t). The options having shortest maturity have been chosen in order to
minimize the noise caused by maturity mismatch, as these forecasts are to be
used as one-week ahead ex-ante forecast of volatility. Notably, the IVs (as an
ex-ante forecast for the next week) have been calculated at the close of trading
on the Friday, which, at the same time, marks the end of the rolling samples
for the estimation of ARCH models. From the options data, a series of 49
implied volatility forecasts of one-week-ahead volatility could be examined for
call as well as put option unlike the total number of 53 forecasts from ARCH
models, as four contracts were lost in filtering the data on specified moneyness
and liquidity criteria. This approach is similar to that adopted by Day and
Lewis (1992) and most of the other studies carried out later on (e.g. Claessen
and Mittinik, 2002).
Having estimated the ex-ante one-week ahead forecasts of volatility from
select ARCH and implied volatility models, the ex-post calculation of realized
volatility for the forecasted weeks has been carried out. Also, like ARCH
models, a total number of 53 one-week ahead realized volatilities were calculated
ex-post. The realized volatilities so calculated have been used as a benchmark
for assessing the forecasting ability of different volatility models under
consideration. The realized volatility for one week represents the average of
daily volatilities for all the trading days of the week. The formula used for
calculation of the realized volatility is given in Eq. (6.9).
n
1
VRt =
n ∑r +
i=1
2
t i (6.9)
 Informational Efficiency of Implied Volatilities of S&P CNX Nifty Index Options ● 143

where, VRt is the realized average variance for the week at time ‘t’, r2t + i is the
squared return for the ith day calculated ex-post and n represents the number
of trading days in the week. The volatility so arrived has been annualized
multiplying by 252 (i.e. number of trading days in a year).
Further, mean error (ME), mean absolute error (MAE) and mean squared
error (MSE), the most commonly used criteria for assessing the forecasting
ability of different models, have been used to examine the forecasts from the
four models (i.e. ‘GARCH’, ‘EGARCH’, ‘Implied volatility of call options’
and ‘Implied volatility of put options’) vis-à-vis the realized volatility for
the week. The ME, MAE and MSE are given in Eqs (6.10), (6.11) and (6.12),
respectively.
n
1
ME =
n ∑
t =1
VtF – VtR (6.10)

n
1
MAE =
n ∑V
t =1
t
F
– VtR (6.11)

n
1 2
MSE =
n ∑
t =1
VtF – VtR (6.12)

where, VFt is one-week ahead ex-ante forecast from one of the select volatility
W
models under consideration, specifically, h F , t in the case of ARCH models and
VtI in the case of implied volatility models; n denotes the number of forecasts
made using each method, i.e. 53 in the case of ARCH models and 49 in the
case of implied volatility models; and VRt is the ex-post estimation of one-week
ahead realized volatility.
The results pertaining to the ex-ante forecasting performance of all the
volatility models under consideration are summarized in Table 6.5. The
findings clearly indicate that forecasts from the EGARCH model are the best
amongst all other competing models used to forecast the ex-ante volatility.
Also, the forecast from the GARCH model have shown better results compared
with those based on IVs of call as well as put options as indicated by MAE
and MSE. To sum up, the measures of forecast efficiency in terms of overall
MAE and MSE indicate that IVs have failed to generate considerably good
forecasts compared with those from the ARCH models both in the cases of
call and put options.
Notwithstanding the overall MAE and MSE, the overall ME designates IVs
from the call options as the best ex-ante forecast of future volatility because
it turns out to be the least. However, the conceptual superiority of the other
two measures, namely MAE and MSE, lessens the importance and reliability
of this finding, as the ME is sensitive to the sign of deviation. That is, positive
and negative deviations are neutralized in calculation of ME.
144 ● Derivative Markets in India

The forecasts based on IVs seem to have failed in impounding the ARCH
as well as leverage effect, which probably be assigned as a major reason for
relatively high forecast error compared with ARCH model. This finding, prima-
facie, rejects the hypothesis that the IVs have subsumed all the information
contained in ARCH forecasts, which are based on historical returns data. And,
therefore, the IVs can be designated as informational inefficient as these do
not seem to have impounded relevant information about the future forecasts
compared with ARCH models.
Since the above-mentioned findings are based on the overall measures of
forecast performance, there is always a possibility that such measures might
present a distorted picture on account of the presence of a few outliers in
the data. Therefore, it would be useful to further analyse the data for more
reliable results using different percentiles to test the consistency in forecasting
efficiency of different models under consideration. The percentiles are reported
in Table 6.5.
The results in terms of percentiles corroborate the above finding that
forecasts from the EGARCH model are the best as the first three quartiles (25th,
50th and 75th percentiles) turn out to be the least for this model compared with
all the other three competing models. That is, in 75% cases, the forecasts from
the EGARCH models have outperformed the forecasts from the other models
and, therefore, reflects the consistency of EGARCH ex-ante forecasts.
Moreover, if we look at all the percentiles calculated, the forecasts from the
EGARCH model have outperformed the forecasts from the implied volatility
in a total number of 85% cases, i.e. first 75% cases plus 10% cases from 85th
to 95th percentiles. In other words, the forecasts from the IVs of call options
(the next best model) have outperformed the EGARCH forecast only in 15%
of the total cases analysed. Likewise, the forecasts from the GARCH model
have outperformed the forecasts from the IVs of call options in a total number
of 65% of the cases analysed, i.e. first 50% cases plus 15% cases from 85th to
100th percentiles. Thus, the forecasts from implied volatility of call options
have outperformed the GARCH forecasts in 35%s of cases analysed.
In contrast, the forecasts from ARCH models have outperformed the
forecasts from the IVs of put options in a total number of 90% cases. That is,
the forecasts based on IVs of put options could outperform the forecasts from
the ARCH models only in 10% of the cases.
In view of the above findings, it would be appropriate to conclude that IVs
do not have much information about the futures volatility. These findings
corroborate the earlier finding (based on the overall ME, MAE and MSE) that
the IVs are informational inefficient. However, one of the plausible arguments
in favour of implied volatility could be the maturity mismatch. That is, IVs of
the options having 8–30 days to maturity have been used to forecast one-week
ahead volatility. Therefore, it would be appropriate to analyse the options that
 Table 6.5 Comparative Forecasting Performance of Implied Volatility Models for Call as well as Put Options Having 8–30 Days to Maturity
vis-à-vis Select Conditional Volatility Models in Forecasting One-Week-Ahead Volatility, July 2006 to June 2006

Forecasting models
Performance
GARCH(1,1) EGARCH(1,1) Implied variance call options Implied variance put options
measurement
ME MAE MSE ME MAE MSE ME MAE MSE ME MAE MSE
Overall
0.00380 0.03083 0.00188 –0.00256 0.02700 0.00172 0.00156 0.03208 0.00199 0.03646 0.04953 0.00320
Percentiles
25 0.00229 0.01065 0.00012 –0.00311 0.00859 0.00007 –0.00750 0.01501 0.00023 0.02104 0.03332 0.00111
50 0.01530 0.02024 0.00041 0.00964 0.01658 0.00027 0.01567 0.02469 0.00061 0.04053 0.04508 0.00203
75 0.02484 0.03977 0.00159 0.01875 0.02851 0.00082 0.03054 0.03541 0.00126 0.05925 0.06447 0.00416
80 0.02667 0.04830 0.00234 0.02131 0.03654 0.00135 0.03339 0.03635 0.00132 0.06465 0.06679 0.00446
85 0.03492 0.06916 0.00480 0.02505 0.04998 0.00250 0.03482 0.03860 0.00149 0.07036 0.08225 0.00677
90 0.03981 0.07563 0.00573 0.02890 0.07250 0.00525 0.03606 0.08018 0.00643 0.08242 0.09462 0.00895
95 0.05658 0.10436 0.01096 0.03752 0.11309 0.01281 0.03860 0.12754 0.01644 0.10205 0.10204 0.01041
100 0.07065 0.13849 0.01918 0.05057 0.14231 0.02025 0.04214 0.14248 0.02030 0.12827 0.12827 0.01645
Note: In the table, ME, MAE and MSE denote mean error, mean absolute error and mean squared error, respectively.
Informational Efficiency of Implied Volatilities of S&P CNX Nifty Index Options
●
145
146 ● Derivative Markets in India

are close to the one-week forecast horizon (in terms of maturity) to assess the
true informational efficiency of IVs in forecasting the future volatility. For the
purpose, the IVs of the options having 8 days to maturity have been used to
compare the ex-ante one-week ahead forecasting ability of different volatility
models under consideration. The results in this regard are summarized in
Table 6.6.
The results clearly indicate that the forecasts from the IVs of call options
turn out to the best, as the overall ME, MAE and MSE are the least amongst all
the competing models. However, the percentiles corroborate that the forecasts
from the EGARCH model still remain the best forecasts, as the MAE and MSE
for the first three quartiles (for 75% of the cases) are the least among all the
models. That is, the percentage of forecasts from EGARCH that outperformed
the implied volatility forecasts from the call options has reduced from 85% to
75% and can be attributed to the correction of maturity mismatch.
At the same time, the forecasts from the IVs have shown considerable
improvement vis-à-vis GARCH forecasts when corrected for maturity
mismatch. The forecasts from the IVs of call options have outperformed all
the forecasts from the GARCH model; the percentage of GARCH forecasts
outperforming the forecasts based on the IVs of call options has reduced from
65% to 0. In view of these findings, it would be appropriate to infer that the
informational inefficiency in the IVs of call options can, to a marked extent,
be attributed to the maturity mismatch.
In sum, it can be concluded that the forecasts from the IVs of the call
options impound all the information contained in the GARCH forecast. In
other words, the IVs of call options do contain relevant information about
the future volatility; however, these failed to capture the leverage effect as
the forecasts from the EGARCH models still outperformed implied volatility
in 75% of the cases.
Unlike the IVs of call options, those of put options have shown no
improvement in their forecasting ability. It is evident from the fact that the
forecasts of put options from the ARCH models have outperformed the
forecasts based on IVs in 95% of the cases. This finding corroborates the earlier
finding that the IVs of put options are clearly informational inefficient and do
not contain much relevant information for forecasting the future volatility.

6.5.4 Comparison of Call and Put Options

The ‘in-the-sample’ analysis has revealed that the IVs of call as well as put
options are informational inefficient. However, a comparative examination
of IVs demonstrates that IVs of call options have impounded higher amount
of information available in the historical returns compared with those of put
options. It is evident from the fact that the coefficient of IVs in the GARCH
variance equation is 0.4354 and 0.1107 for call and put options, respectively.
 Table 6.6 Comparative Forecasting Performance of Implied Volatility Models for Call as well as Put Options Having 8 Days to Maturity
vis-à-vis Select Conditional Volatility Models in Forecasting One-Week-Ahead Volatility, July 2006 to June 2006
Forecasting models
Performance GARCH(1,1) EGARCH(1,1) Implied variance call options Implied variance put options
measurement
ME MAE MSE ME MAE MSE ME MAE MSE ME MAE MSE
Average
0.00380 0.03083 0.00188 –0.00256 0.02700 0.00172 0.01964 0.01964 0.00053 0.04958 0.04958 0.00356
Percentiles
25 0.00229 0.01065 0.00012 –0.00311 0.00859 0.00007 0.00947 0.00947 0.00009 0.02378 0.02378 0.00061
50 0.01530 0.02024 0.00041 0.00964 0.01658 0.00027 0.01673 0.01673 0.00029 0.04263 0.04263 0.00182
75 0.02484 0.03977 0.00159 0.01875 0.02851 0.00082 0.03495 0.03495 0.00123 0.06157 0.06157 0.00381
80 0.02667 0.04830 0.00234 0.02131 0.03654 0.00135 0.03536 0.03536 0.00125 0.07882 0.07882 0.00653
85 0.03492 0.06916 0.00480 0.02505 0.04998 0.00250 0.03569 0.03568 0.00127 0.10199 0.10199 0.01044
90 0.03981 0.07563 0.00573 0.02890 0.07250 0.00525 0.03614 0.03614 0.00131 0.11997 0.11997 0.01455
95 0.05658 0.10436 0.01096 0.03752 0.11309 0.01281 0.03635 0.03635 0.00132 0.12827 0.12827 0.01645
100 0.07065 0.13849 0.01918 0.05057 0.14231 0.02025 0.03635 0.03635 0.00132 0.12827 0.12827 0.01645
Note: In the table, ME, MAE and MSE signify mean error, mean absolute error and mean squared error.
Informational Efficiency of Implied Volatilities of S&P CNX Nifty Index Options
●
147
148 ● Derivative Markets in India

Similarly, the respective coefficients turn out to be 0.1654 and 0.0890 for the
EGARCH variance equation. These findings denote that the IVs of call options
are more informational efficient compared with those of put options.
Further, the forecasts based on IVs of put options have been outperformed
by the forecast of all the other models under consideration (Tables 6.5 and
6.6). Notably, the forecasts based on IVs of put options have consistently
been outperformed by those based on the IVs of call options, as manifested
in ME, MAE and MSE on aggregate as well as disaggregate basis (in terms of
percentiles). Moreover, the MEs in the case of forecasts from the IVs of put
options have consistently been higher (nearly more than double) compared
with those in the case of call options. This finding (along with the MAE and
MSE), per-se, reveals that the IVs of put options are consistently overstating the
forecasts of future volatility compared with those of call options and signifies
that the IVs of put options are overpriced. The relative overpricing of IVs of put
options can be attributed to the overpricing of the put options, as the call and
put options used for out-of-the-sample analysis have the same characteristics,
viz. trading date, maturity date and strike price. The overpricing of put
options in relation to the corresponding call options is, apparently, indicative
of violation of the put–call parity relationship.
Therefore, the poor performance of IVs of put options vis-à-vis those
of call options, as revealed by in-the-sample as well as out-of the-sample
analysis, can be attributed to the overpricing of put options. Also, it would
be appropriate to conclude that the IVs of put options are more informational
inefficient compared with those of call options. However, such anomalies in
an efficient market should be corrected by arbitrage mechanism, provided the
market microstructure facilitates the arbitrage processes required to correct
such anomalies.
Given the state of market microstructure of Indian securities market
during the period under reference, the relative overpricing of put options can
possibly be attributed to the short-selling constraint. This has been assigned as
one of the major reasons in view of the fact that to exploit such an arbitrage
opportunity, the three courses of action, namely (i) sell the overpriced put,
(ii) buy a corresponding call with the same contract specifications and (iii) sell
the underlying asset(s) short, are required to be taken. In short, relatively lower
informational efficiency of the IVs of put options could be traced to the existing
market microstructure in India during the period under reference.
The major finding of the study that the IVs do not incorporate all the
information available in the historical returns, which signifies the informational
inefficiency of IVs vis-à-vis the volatility estimates from select ARCH models,
is in line with Day and Lewis (1992), Canina and Figlewski (1993), Lamoureux
and Lastrapes (1993) and Pong et al. (2004) for the forecast horizon of one
 Informational Efficiency of Implied Volatilities of S&P CNX Nifty Index Options ● 149

day and one-week. However, at the same time, the results of the study are
opposite to some studies that found IVs to be informational efficient. These
include Latane and Rendleman (1976), Chiras and Manaster (1978), Gemmill
(1986), Shastri and Tandon (1986), Scott and Tucker (1989), Xu and Taylor
(1995), Fleming, J. (1998) and Claessen and Mittnik (2002).

6.6 CONCLUDING OBSERVATIONS

The present study has attempted to assess the informational efficiency of IVs
vis-à-vis volatility estimates from select conditional volatility models, namely
GARCH(1,1) and EGARCH(1,1). In the study, the IVs have been analysed for
‘in-the-sample’ as well as ‘out-of-the-sample’ data. The results of the analysis
reveal that the IVs have failed to capture all the information available in the
historical returns, and the forecasts based on the IVs are inferior to those based
on the ARCH models. Though the IVs have been found to be informational
inefficient for both call and put options, the IVs of call options, when corrected
for maturity mismatch, have shown marked improvement in the quality of
forecasts. This is borne out by the fact that these successfully impounded all
the information available in the forecasts based on GARCH model in case of
‘out-of-the-sample’ analysis, though it failed to incorporate the leverage effect,
which remained the basic reason for the best quality forecasts from EGARCH
model. In contrast, the put options market has shown comparatively higher
inefficiency both for in-the-sample and out-of-the-sample analysis.
In short, the implied volatility of call options does have relevant information
about the future volatility; however, it fails to capture all the information
available in the historical returns (especially in terms of leverage effect). On the
contrary, the implied volatility of put options does not show any improvement
in the quality of its forecasts of future volatility even when corrected for
maturity mismatch. Therefore, the implied volatility of put options has
remained more informational inefficient compared with call options.
The relative overpricing of implied volatility of put options compared with
call options can be attributed to the short-selling constraint in Indian securities
market under reference, as the arbitrage mechanisms that corrects such
anomalies requires the underlying assets to be sold short. Notwithstanding
the constraints in terms of market microstructure of Indian securities market,
such anomalies are expected to hinder the performance of options on its well
identified functions, namely risk hedging, price discovery and the resultant
of these two, i.e. the efficient allocation of funds.
The informational inefficiency of IVs is indicative of mispricing of options
contracts in Indian securities market and puts a question mark on performance
of the options market.
150 ● Derivative Markets in India

END NOTES
1. The non-negativity constraints on the coefficients of an ARCH variance
equation rule out the possibility of negative coefficients, as the values
of the coefficients are restricted to zero or a positive value (may or may
not be specified explicitly). This, in turn, does away with the possibility
of a negative variance estimate.
2. Generalized Error Distribution (GED) is a theoretical distribution, which
is suitable for the data which have fat-tails and high peakedness.
3. It represents a model where no constraint has been imposed on the
value of coefficients in the model.
4. It represents a model where one or more coefficients have been
constrained to some value, e.g. zero.
7 C H A P T E R

Survey Analysis and

Findings

7.1 INTRODUCTION
This survey attempts to gauge viewpoint of trading member organizations/
brokers on the state of pricing of options contracts in Indian derivatives
market and reforms in terms of regulation and educational initiatives for the
betterment of the market. For the purpose, a questionnaire has been prepared
to seek responses/perception of trading member organizations on the issue. The
survey was conducted amongst the branch managers and research analysts of
brokerage firms based at Delhi and Mumbai. The survey is expected to provide
a useful insight primarily for two reasons: (i) the brokerage firms play a central
role in capital creation in an economy as they facilitate trading of financial
securities/assets (including derivatives securities), and (ii) derivatives markets
is new in India as it took off in June 2000 only, and therefore, the understanding
of the market amongst such intermediaries needs to be assessed.
In the survey, the viewpoint of the respondents has been captured on five
major aspects. These are: (i) level of participation and usage of the options market,
(ii) awareness and use of models for options valuation, (iii) understanding
of put–call parity relationship, (iv) correctness of options pricing, its impact
and existence and exploitability of arbitrage opportunities and (v) need of
regulations and educational initiatives for the betterment of the market.
The rest of the chapter has been organized in four sections. The Section 7.2
enumerates survey methodology. Profile of respondents has been summarized
in Section 7.3. Section 7.4 contains analysis and empirical evidences. The
chapter ends with concluding observations in Section 7.5.

7.2 SURVEY METHODOLOGY

7.2.1 Questionnaire Development, Scales Used in the Questionnaire

and Identification of the Target Population
In the survey, questionnaire has been used as a major tool for collecting
required information from the respondents. The questionnaire was developed
152 ● Derivative Markets in India

based on the experts’ opinion and an earlier survey on derivatives in Indian

securities market by Srivastava et al. (2008). Expert opinion was sought
from five eminent academician (including the head of the Committee on
Development of Derivatives Market in India) and five branch managers of
different trading member organizations in Delhi, actively dealing with the
derivatives segment. The questionnaire is based on two measurement scales,
namely nominal and interval scales, for the measurement of the responses.
Since the purpose of the survey has been to gauge the perception of the
market participants of the pricing of options, the questionnaire has been
administered amongst the trading member organizations. The trading member
organizations play a crucial role in the financial markets, as they not only
facilitate trading in the market but also participate actively in the trading. The
understanding of derivatives market is crucial amongst such organizations,
as they facilitate trading for retail (including high net worth individuals) and
institutional investors (through proprietary trading). Moreover, the investors
seek the advice of such organizations regarding their investment decisions
from time to time. Therefore, based on the active participation and crucial
advisory role, we have selected trading member organizations as the target
population for the purpose of the survey to gauge the market perception on
the pricing of options contract in Indian derivatives market. For the purpose,
a list of total number of trading member organizations active in derivatives
market as on September 2007 was retrieved from the website of Securities and
Exchange Board of India (SEBI).

7.2.2 Pre-Testing of Questionnaire (Reliability and Validity)

The questionnaire was firmed up after incorporating suggestions and
improvements from various practitioners and academics, and the validity and
reliability of the instrument have been tested. For the purpose, it has been pre-
tested on five branch managers from different trading member organizations
based at Delhi and five eminent academics.
Reliability and validity refer to the ability of the instrument to measure
consistently what it proposes to measure. The face and/or content validity of
the questionnaire have been ensured by discussing the responses with experts
on the subject. In addition to this, the reliability of the instrument was tested
for the related questions, which have been measured on interval scale. The
reliability of the related questions that constitute a construct has been tested
using Chronbach’s alpha. The coefficient, i.e. the Chronbach’s alpha, turned
out to be more than 0.66 for all related questions. This reasonably good
value of Cronbach’s alpha for all related questions confirms reliability of the
questionnaire.
 Survey Analysis and Findings ● 153

7.2.3 Sample Size Determination

Moreover, based on the responses from the pilot survey, an attempt has
been made to statistically determine the sample size. The statistical approach
adopted to determine the sample size is based on traditional statistical
inference as mentioned in Malhotra (2005). In this approach, the desired
precision level is specified in advance. The approach is based on creating
confidence intervals around the mean.
The sample size determination under this approach requires five steps as
mentioned in Malhotra (2005). These steps are:
(i) Specifying the level of precision—it represents the maximum permissible
difference between the sample and population mean of the characteristics
of interest. The level of precision can be specified in absolute terms as
well as in relative terms (in per cent). In the study, it has been set at the
commonly used level of 5%.
(ii) Specifying the level of confidence—The level of confidence in the present
study has been set at 95%.
(iii) Determining the z value associated with the specified level of
confidence.
(iv) Estimation of the sample mean and standard deviation of the
characteristics of interest in case the population mean and standard
deviation are unknown and
(v) Determining the sample size using the formula given in Eq. (7.1).
C 2 z2
n= (7.1)
R2
where, C is the coefficient of variation, (s/m), z is the value of the standardized
normal variate at 95% level of confidence and R represents the precision level/
maximum permissible difference in percentage terms.
Based on the 10 responses from the pilot survey, the mean and standard
deviation for all the characteristics of interest was estimated and an average
of mean of all the characteristics of interest along with the average variance
was determined. The aggregate mean and standard deviation for the pilot
data have been 3.62 and 0.7575, respectively. Finally, we arrived at the sample
size of 67 for the 95% confidence level and 5% level of precision, based on the
aggregate mean and standard deviation determined from the data.

7.2.4 Sample Selection/Sampling Technique Used

We have used a two-stage cluster sampling technique as we are required to
gauge the perception of market participants/trading member organizations
based at Delhi and Mumbai. The geographical cluster/area sampling has
been applied in view of the fact that the population to be surveyed fits the
154 ● Derivative Markets in India

requirement for such a sampling technique, i.e. externally homogenous and

internally heterogeneous groups/clusters. The respondents of the survey
form ‘externally homogeneous’ geographical clusters in terms of the cities they
are located at and are ‘internally heterogeneous’ based on their different trading
volumes within a given city.
The two stages of the sampling technique are (i) choosing clusters of the
populations to be surveyed and (ii) choosing sampling units from within the
selected clusters. Amongst all the clusters of the target population, we have
chosen to study the two selected clusters (Delhi and Mumbai). The survey
has been focused on the trading member organization situated at Delhi and
Mumbai only, as the majority of trading member organizations are located at
these two places. In addition, an earlier survey on derivatives by Srivastava et
al. (2008) revealed that majority of responses, i.e. 78% of the total responses,
were from these two cities. This then constitutes the rationale of limiting the
survey to the trading member organizations of Delhi and Mumbai only.
Further, simple random sampling has been used at the second stage of
sampling in order to select the sampling units from within each cluster. This
has been done in view of the fact that it was not feasible to survey the whole
population of select clusters owing to the budgetary constraints. For the purpose,
we have generated 100 random numbers within the range of 1–172 for Delhi.
Likewise, 250 random numbers have been generated within the range of 1–345
for the selection of trading member organizations in Mumbai. Based on these
numbers, a total number of 350 respondents have been surveyed; 100 from
Delhi and 250 from Mumbai.

7.2.5 Questionnaire Administration and Data Collection

The pilot-tested questionnaire was sent to a total number of 100 and 250 trading
organizations out of the total population of 172 and 345 trading member
organizations based at Delhi and Mumbai, respectively. Subsequently, a
reminder was sent through email after one week of the date the questionnaire
was mailed though postal mail. In total, 350 questionnaires were mailed to the
randomly selected trading member organizations based at Delhi and Mumbai,
out of the total population of 517 organizations. Moreover, all such trading
member organizations were also sent a questionnaire through email.
The initial response was very poor as only 10 trading members responded.
Two reminders were sent to the remaining respondents through email mode
with the interval of one month and two months, respectively, from the date of the
first reminder. In addition, a total number of 20 trading member organizations
were surveyed by personally visiting to the offices of organizations based at
Delhi. For the purpose, the researcher sought appointments through email
 Survey Analysis and Findings ● 155

and phone calls. As a result, in the time span of approximately 5 months from
March 2008 to August 2008, a total number of 64 responses were received. The
responses include 31 responses from Delhi and 33 from Mumbai. Out of all the
31 responses from Delhi, 19 were collected through personal visits and rest of
the responses were received though postal mail (9) and email (3). Similarly,
all the 33 responses from Mumbai were received through postal mail (26) and
email (7). This amounts to 12.38% of the total population of trading members
active in derivatives segments.
Though the response rate is less than one-fifth (18.29%, i.e. 64 out of 350), it
should not be considered as low/poor response in view of the busy schedule
of executives of trading member organizations. Another reason for such a
response rate is that business organizations normally consider information
related to financial matters very sensitive and confidential. However, the
response level should be borne in mind while interpreting the results/findings
of the survey.

7.3 PROFILE OF RESPONDENTS IN TERMS OF THEIR BACKGROUND

AND TRADING VOLUME
As mentioned earlier in the methodology section, the respondents of the survey
belong to the two major clusters/ hubs of trading for derivatives securities,
namely Delhi and Mumbai. The number of respondents from Delhi and
Mumbai were 31 and 33, which account for 48% and 52%, respectively, of
the total number of respondents. Almost equal percentages of respondents
from the two clusters represent a balanced sample, which assigns equal
importance to the opinion of the respondents from both Delhi and Mumbai.
The geographical background of the respondents has been presented in
Fig. 7.1.

Figure 7.1 Geographical location of the respondents

In addition, profile of the respondents based on their trading volume (based

on the notional value of contracts) in derivatives (F&O) segment has been
depicted in Fig. 7.2. The results enumerate that majority of the traders, i.e.
156 ● Derivative Markets in India

51% of total respondents had average daily turnover (notional value) between
` 25 and 100 crore, one-fourth of the respondents had daily trading volume of
more than ` 100 crore, and nearly one-fourth (24%) of the respondents belongs
to the category having a daily turnover of less than ` 25 crore.

Figure 7.2 Average daily turnover (in `) of the respondents in

Futures & Options (F&O) segment

7.4 ANALYSIS AND EMPIRICAL RESULTS

The analysis of the responses has been classified into five subsections relating
to the viewpoint of trading member organizations on pricing of options
contracts in Indian derivatives market and need of regulation and educational
initiatives for the betterment of the market. These subsections represent the
five dimensions proposed to be measured by the survey.

7.4.1 Participation in Options Market and Its Usage

In order to examine the popularity of options vis-à-vis other derivative
products in Indian securities market, the respondents were asked on the share
of options in their total trading volume relating to derivatives. The results in
this respect have been depicted in Fig. 7.3. The results of the survey reveal that
the vast majority of participants (79%) have been found to have a relatively
meagre share (up to 20%) in the options compared with other derivative
products in the portfolio.
The finding is corroborated by the fact that more than one-third (35%)
of the respondents had less than 10% of their trading volume in options;
majority of which were the respondents having less than 5% proportion
in derivatives trading devoted to options. In addition, a major segment of
respondents representing 44% of the participants have invested on an average
10–20% in options of their trading volume relating to derivatives segment.
 Survey Analysis and Findings ● 157

Only 16% of the respondents have included a reasonably good proportion of

options in their trading volume from derivatives segment by assigning 20–50%
weightage to the options contracts in their portfolio of derivatives securities;
just 5% of total respondents have shown very high interest in options as they
have more than 50% of their activity in derivatives market through options
market.

Figure 7.3 Proportion of options trading in the total trading

volume relating to derivatives

It would be reasonable to infer from the above findings that the respondents
of the survey have accorded relatively less weightage to options; their major
interest seems to be in futures contracts. The finding for marked preference
to the futures is in conformity with an earlier survey in the Indian derivatives
market by Srivastava et al. (2008).
It was also of interest to ascertain the purpose of taking a position in options
market. The results have been presented in Fig. 7.4. An overwhelming majority

Figure 7.4 Purpose of taking a position in options market

158 ● Derivative Markets in India

(95%) of the respondents have been using this market primarily for hedging, in
conjunction with other objectives, e.g. speculation and arbitrage. Speculation
(75% of respondents), along with other purposes, has been the second most
preferred objective; however, arbitrage emerged as the least preferred objective
for entering the market. The findings of survey, to a marked extent, are in line
with the report of the L.C. Gupta Committee (1998) on derivatives market. The
Committee reported that majority of the respondents (70%) aspired to have a
derivatives market to get a proper hedging mechanism for the equity portfolios.

7.4.2 Awareness and Use of Models for Options Valuation

The survey, amongst other objectives, has attempted to gauge the level of
understanding amongst respondents on the important aspects of options
valuation. For the purpose, the respondents of the survey have been asked on
difficulty in valuing options contracts, factors that need to be considered while
valuing options and the valuation models. The results relating to the difficulty
in valuation of options in terms of their comparison with futures, difficulty
in estimating volatility for valuing options and use of implied volatility for
valuing option have been summarized in Table 7.1. The results demonstrate
that the respondents have shown their strong agreement on the difficulty
in valuation of options compared with futures, difficulty in estimating the
volatility and using volatility for valuing options. The results have been found
to be highly significant. The agreement on these aspect of options valuation,
prima-facie, indicate that the respondents are familiar with the major difficulty
in valuing an options contract.
Further, the study has attempted to assess the level of awareness amongst
respondents on options valuations models and their propensity to use such
models for the valuation of options. The results on awareness of valuation
models are depicted in Fig. 7.5.

Figure 7.5 Awareness of options valuation models amongst respondents of the survey
 Table 7.1 One-Sample ‘T’ Test for The Statistical Validation of Agreement/Disagreement of Respondents on the Some Aspects Relating to Valuation
of Options
Test value = 03
95% confidence interval of
Statement Number X s Mean
t df Sig. the difference
difference
Lower Upper
Valuation of options is more
64 4.19 0.687 13.83 63 0.000 1.188 1.02 1.36
difficult than that of futures
Volatility is most difficult to be
64 3.69 0.990 5.56 63 0.000 0.688 0.44 0.93
estimated
Implied volatility can be used to
64 3.89 0.620 11.49 63 0.000 0.891 0.74 1.05
value the options
Survey Analysis and Findings
●
159
160 ● Derivative Markets in India

The results on awareness of valuation models clearly indicate that the

Black–Scholes (BS) model has been found to be the most popular amongst
the respondents compared with the other two models, namely Binomial-Tree
(BT) model and Hull and White (HW) model. It is corroborated by the fact
that two-third (67%) of the respondents have been found to be aware of the BS
model; the figure has been much lower for other models. For instance, while
nearly one-tenth of the respondents were aware of BT model, HW model was
known to just 6% of the respondents.
It is pertinent and revealing to note that 30% of the respondents were not
aware of any model. Equally notable finding is that nearly one-half of the total
respondents have not been using any model despite the fact that 70% of the
respondents are aware of the valuation models.
It was of interest to know from the respondents which model/models of
valuation (amongst the three mentioned earlier) have been used to value the
options. The results on the use of specific model are summarized in Table 7.2;
it is interesting to note that the BS model is popular amongst the respondents
as nearly half of the respondents (49%) have been found to be using the BS
model. In sharp contrast, just 3% of the respondents have been using BT model
and no respondent used HW model for valuing options. In addition to this,
the consistency of awareness and use of models has been examined using
cross tabulation of awareness and use of the specific models. The results are
summarized in Table 7.2.
The results reveal that the 49% of the respondents who are aware of the BS
model are actually using it. At the same time, 3% of the respondents who were
aware of the BT model have been valuing options using this model. However,
6% of the respondents were those who claimed to be aware of all the models
but actually have not been using any model.
Table 7.2 Cross Tabulation of the Responses on Awareness
and Use of Options Valuation Models
Which of the model do you use
None of Total
BS BT
the above
BS model Count 29 0 8 37
% of Total 45.3% 0.0% 12.5% 57.8%
BT model Count 0 2 0 2
Awareness of % of Total 0.0% 3.1% 0.0% 3.1%
the valuation None of the above Count 0 0 19 19
models % of Total 0.0% 0.0% 29.7% 29.7%
BS, BT and HW Count 0 0 4 4
% of Total 0.0% 0.0% 6.3% 6.3%
BS and BT Count 2 0 0 2
% of Total 3.1% 0.0% 0.0% 3.1%
Count 31 2 31 64
Total
% of Total 48.4% 3.1% 48.4% 100.0%
 Survey Analysis and Findings ● 161

Figure 7.6 Percentage of respondents considering important factors for valuing

index and stock options

Figure 7.6 contains the factors considered by those respondents who have
been using BS and BT models in valuing options. The results depicted in the
figure are revealing, as the respondents who are using BS and BT models in
pricings options have not considered all the factors reckoned by these models in
valuing options. These models, in general, include six variables, namelystrike
price, spot price, time to maturity, volatility of underlying asset, interest rate
and dividend yield/absolute amount of dividends, to arrive at the correct
price for the options.
The vast majority of respondents (80%) have considered the first four factors
in valuing index or stock options; it is surprising to note that the last two
factors, namely interest rate and dividend yield, have not been used widely
in valuing the options. Amongst the four most considered factors, volatility
of underlying asset has been indicated as the most important factor, as all
the respondents considered it in valuing the options. As far as interest rate
is concerned, only 52% and 61% of the respondents have used it in valuing
index and stock options, respectively; the figure is much lower at 23% and 35%,
respectively, pertaining to the use of dividend yields for valuation purpose.
The finding that interest rate is one of the two least considered factors for
valuing options amongst respondent is in line with Srivastava et al. (2008). It
may be noted that the consideration of dividend yield in valuation would not
affect the correct prices much; however, the valuation using these models is
not possible without considering the interest rate, as it is a necessary input
required to determine options price.
In sum, it would be reasonable to conclude that the ‘true’ percentage of the
respondents who are actually using the ‘correct’ valuation models is lower
than that noted earlier (52%). The actual percentage represents the respondents
who reckoned interest rate also (along with other variables) as an important
factor for valuing options. Based on such a correct measure, nearly one-fourth
162 ● Derivative Markets in India

of the respondents are actually using valuation models (BS and BT) for valuing
index options. Likewise, a marginally higher percentage of respondents have
been found to be actually using valuation models for pricing options.
As far as the method used for estimating volatility is concerned, half of
the respondents have been found to be using standard deviation method
(Fig. 7.7). In contrast, only 16% of the respondents have been using moving
average models; surprisingly, the remaining one-third (34%) of the respondents
were not using any model to estimate the volatility.

Figure 7.7 Percentage of respondents using specific model for estimating volatility

7.4.3 Understanding of Put–Call Parity Relationship

This section addresses the issue relating to the level of understanding of put-
call parity (PCP) relationship amongst the respondents of the survey. The PCP
condition denotes relationship between the price of a call and a corresponding
put option with the same contract specification. The understanding of PCP
relationship is important as it helps to identify the arbitrage opportunities
(mispricing signals), which exist on account of the violation of this relationship.
In other words, it helps to spot the arbitrage opportunities in terms of
underpricing/overpricing of a call (put) option relative to a put (call) option
with same contract specifications.
For the purpose, the questionnaire included three questions on the
understanding of PCP relationship. The first is a straight question asking the
participants whether they are aware of PCP; the next two questions have been
asked to confirm whether they know the concept well.
The responses to the first question on PCP relationship are depicted in
Fig. 7.7. The results indicate that nearly two-thirds of the respondents are aware
of the PCP relationship. It is a matter of concern that nearly one-third of the
respondents has shown lack of understanding of PCP relationship. However,
 Survey Analysis and Findings ● 163

there is a possibility that the participants might have good understanding of

the concept; they might not be aware of the terminology, i.e. PCP relationship.
In order to gauge the true level of awareness of the PCP relationship amongst
the participants, we need to analyse these results in light of the responses to
the other two questions on the parity relationship.
The responses on the next two questions have first been analysed using ‘one-
sample t-test’ in order to identify if the mean responses indicate a statistically
significant agreement, disagreement or neutral opinion of the respondents
on the PCP relationship. For the purpose, the null hypotheses tested are that
the mean score m = 0 against the alternative m > 3 and m < 3 in the first and
second question, respectively. The results on next two questions on PCP are
summarized in Table 7.3.
The results indicate that the null hypothesis is rejected at 5% level of
significance in the case of second (first in Table 7.3) question on PCP; the
associated empirical value of t statistics is 3.347, which indicates that the
agreement on the statement has been found to be statistically significant. In
contrast, the null hypothesis could not be rejected at 5% level of significance
in the case of the third (second in Table 7.3) question on PCP. It indicates
that the respondents had a mixed opinion on the third question, where a
disagreement was expected if they knew the concept well. The results of the
third question are in contradiction with those in the case of second question.
Thus, it is apparent that there is lack of proper understanding of the PCP
relationship amongst the participants of the survey.

Figure 7.8 Percentage of responses on the understanding of PCP relationship

Since the first direct question has revealed that nearly one-third (34%)
of the participants are not aware of the PCP relationship and the next two
questions have shown contradictory results on the PCP relationship, it would
be reasonable to conclude that a significant proportion of the total respondents
lack correct understanding of the concept.
This issue has further been examined by using cross tabulation of responses
to all the three questions on PCP relationship. The results are summarized
in Table 7.4. It is pertinent to note that nearly one-third (32%) of the
 164
●

Table 7.3 One-Sample t-test for the Responses on the PCP Relationship
Test value = 03
95% confidence interval of
Statement Number s
Derivative Markets in India

X t Mean the difference

Df Sig.
difference
Lower Upper
Price of a European call (put)
can be calculated from put (call)
64 3.50 1.195 3.347 63 0.001 0.500 0.20 0.80
option with same contract
specifications.
There is a need to calculate price
of a European call and put (with
64 3.25 1.195 1.673 63 0.099 0.250 −0.05 0.55
same contract specifications)
separately.
Note: In the table, X, s and df denote mean, standard deviation and degree of freedom, respectively.
 Table 7.4 Cross Tabulation of the Responses on all the Three Questions Asked on PCP Relationship
Question/Statement No. 1 Question/Statement no. 2 Question/Statement no. 3
There is a need to calculate price of call and put
Awareness of put call Total
(with same contract specifications) separately
parity
1 2 3 4 5
0* Count 0 0 2 2
% of Total .0% .0% 9.1% 9.1%
Price of a call(put) can be calculated 2 Count 0 0 5 5
No from put(call) option with same % of Total .0% .0% 22.7% 22.7%
contract specifications 3 Count 0 1 4 5
% of Total .0% 4.5% 18.2% 22.7%
4 Count 7 0 3 10
% of Total 31.8% .0% 13.6% 45.5%
Count 7 1 14 22
Total
% of Total 31.8% 4.5% 63.6% 100.0%
1 Count 0 0 0 0 2 2
% of Total .0% .0% .0% .0% 4.8% 4.8%
2 Count 0 0 0 1 3 4
Price of a call(put) can be calculated % of Total .0% .0% .0% 2.4% 7.1% 9.5%
from put(call) option with same 3 Count 0 0 1 4 0 5
Yes contract specifications % of Total .0% .0% 2.4% 9.5% .0% 11.9%
4 Count 0 9 0 9 3 21
% of Total .0% 21.4% .0% 21.4% 7.1% 50.0%
Count 3 6 1 0 0 10
5
% of Total 7.1% 14.3% 2.4% .0% .0% 23.8%
Survey Analysis and Findings

Count 3 15 2 14 8 42
Total
●

% of Total 7.1% 35.7% 4.8% 33.3% 19.0% 100.0%

*
165

represents those respondents who responded to the question/statement no. 1 and 3 but did not respond to question/statement no. 2.
166 ● Derivative Markets in India

participants who responded negatively to the very first question (of

understanding PCP relationship) have answered correctly to the next two
questions. The finding is revealing and indicates that such respondents
actually knew the parity relationship; perhaps, they responded to the first
question negatively as they were not aware of the terminology (PCP) used
for the relationship.
In addition, an equally revealing finding of the survey is that only 45%
(inclusive of 2.4% of participants responded positive to the first question and
had no opinion on the second question) of the participants who responded
positively to the very first question on PCP relationship could respond
consistently to the next two questions relating to the parity relationship.
However, a major segment of such respondents (i.e. 55%) has been inconsistent
in responding to the other two questions. In short, nearly one-third
(55% ¥ 66% = 36%) of such respondents do not have a correct understanding
of the relationship.
The finding is notable as it seems that a major chunk of respondents
did not understand the concept well. This finding has serious implications for
the pricing efficiency of options in India, as the PCP relationship helps to identify
the pricing anomalies/arbitrage opportunities relating to the relative price of
call and put options and, in turn, helps to restore equilibrium in the market.

7.4.4 Correctness of Options Pricing, Its Impact and Existence and

Exploitability of Arbitrage Opportunities
Another important objective of the survey has been to gauge the opinion of the
respondents on correctness of options pricing, its possible impact on the core
functions that the options market is expected to perform and the existence and
exploitability of arbitrage opportunities. The results regarding viewpoint of the
respondents on the state of options pricing and its direction (i.e. underpricing
or overpricing) have been depicted in Figs 7.9 and 7.10.

Figure 7.9 Opinion of respondents on correctness of options prices

 Survey Analysis and Findings ● 167

Figure 7.10 Opinion of respondents on underpricing/overpricing of options

The results depicted in Figs 7.9 and 7.10 clearly demonstrate that nearly
two-fifth (39%) of the respondents have felt that the options in Indian securities
market are incorrectly priced, whereas more than one-third (34%) of the
respondents have believed that these are correctly priced. The remaining
one-fourth (27%) of the respondents showed their inability to judge the state
of options pricing in the market.
On the direction of mispricing of options contracts, only 9% of the
respondents opined that options in Indian securities market are underpriced
when asked for the direction of mispricing. At the same time, a major segment,
i.e. 34% of the respondents, could not say anything on the issue. Majority of
the respondents, who represents 57% of the participants, were of the opinion
that options are not underpriced, i.e. these could either be correctly priced
or overpriced. In order to deduce the percentage of respondents who have
believed that options are overpriced, the responses on these two dimensions
of options pricing (namely viewpoint on the state of pricing and direction of
mispricing, if any) have been cross tabulated. The results are demonstrated
in Table 7.5 and Fig. 7.11.
Table 7.5 Correctness of Options Pricing in Indian Market: Trading Members’ Perspective
Category of respondents Percentage
Incorrectly priced and overpriced 24
Incorrectly priced and underpriced 9
Incorrectly priced and no opinion on under/overpricing 11
No opinion on correctness and under/overpricing 22
Correctly priced 34
Total 100

In addition to the findings mentioned earlier in this respect, it follows from

the cross-tabulation that nearly one-fourth (24%) of the respondents have
believed that options were overpriced. Notably, an important segment (33%)
168 ● Derivative Markets in India

of the respondents had no opinion on the state of pricing in the market. In

sum, it can be inferred from the results that more than two-fifth (44%) of the
respondents felt that options in Indian securities market are priced incorrectly
and, therefore, indicates pricing inefficiency in Indian options market.

Figure 7.11 Cross tabulation of respondents on correctness and under/overpricing of options

Also, the survey has attempted to gauge perception of the respondents

regarding the impact of incorrect pricing on the core functions of the options
market, namely risk hedging and price discovery. For the purpose, the
respondents have been asked on the probable impact of mispricing on price
discovery in underlying’s market and hedging efficiency. The majority of
the respondents agreed on the likely impact of such pricing anomalies, as a
significant segment (78%) of the participants perceived that correct pricing
leads to correct price discovery in underlying’s market; an approximately equal
percentage of respondents (80%) felt that incorrect options pricing lessens the
hedging efficiency of such instruments.
The results of t-test corroborate the above findings, as the opinion of
respondents on the impact of mispricing on price discovery and risk hedging
has been found to be highly significant (Table 7.6). The high level of significance
along with the positive value of the t-test statistic confirm the agreement on
these issues, viz. incorrect options pricing results in incorrect price discovery
and decreases the hedging efficiency of options.
Further, we attempted to gauge the viewpoint of respondents on
existence and exploitability of arbitrage opportunities in Indian options
market. A whooping number of respondents (94%) agreed that the arbitrage
opportunities do exist in Indian options market. As far as attempting to gain
from the arbitrage opportunities is concerned, the results (summarized in
 Table 7.6 One-Sample ‘t’ Test for the Statistical Validation of Agreement/Disagreement of Respondents on the Impact of Incorrect Pricing of
Options on Price Discovery and Hedging Efficiency
Test value = 03
95% confidence interval of
Statement Number X s Mean
t df Sig. the difference
difference
Lower Upper
Correct options pricing
64 3.69 1.082 5.083 63 0.000 0.688 0.42 0.96
ensures pricing discovery
Incorrect options pricing lessens
64 4.05 0.881 9.510 63 0.000 1.047 0.83 1.27
hedging efficiency
Survey Analysis and Findings
●
169
170 ● Derivative Markets in India

Fig. 7.12) indicate that more than 80% of the respondents had actually tried
for the arbitrage in the options market. However, amongst these respondents,
more than one-fourth (27%) of the respondents have attempted to gain
from arbitrage quite frequently, majority of respondents (i.e. 65%) have not
attempted frequently, and a meager 8% of such respondents rarely tried such
strategies.

Figure 7.12 Percentage of respondents who attempted to gain from arbitrage

opportunities along with their frequency of such attempts

In an equally important finding, responding to the ease of exploiting

arbitrage opportunities in Indian options market, less than one-third (31%)
of the respondents feels that the opportunities can be easily exploited, as
summarized in Fig. 7.13. In sharp contrast to this, a major segment of the
respondents opine that such opportunities are either difficult to exploit or
cannot be exploited at all.

Figure 7.13 Opinion of respondents on exploitability of arbitrage opportunities

Further, the responses on ease to exploit arbitrage opportunities have been

examined in terms of the frequency of attempts that the respondents make to
exploit such profit opportunities. The results are summarized in Table 7.7. The
findings reveal that approximately 5% of the respondents who never attempted
to gain from such opportunities have opined that the opportunities can easily
be exploited. This may be attributed to the fact that such respondents, perhaps,
do not understand the arbitrage mechanism very well. At the same time, in
 Survey Analysis and Findings ● 171

a natural response, nearly one-sixth (14%) of the respondents who never

attempted to gain from such opportunities assigned difficulty in exploiting it as
the major reason. Similarly, more than two-fifth (41%) of the total respondents
who have attempted to gain from such arbitrage opportunities feel that these
have been difficult to exploit whereas 6% of such respondents believed that
such opportunities have been unexploitable.
Table 7.7 Cross Tabulation of Frequency of Exploiting Arbitrage and Opinion on
Exploitability of Such Opportunities
Ease of exploiting arbitrage
opportunities
Can be Are Total
Can’t be
easily difficult to
exploited
exploited exploit
Count 3 9 0 12
Never
% of Total 4.7% 14.1% .0% 18.8%
Count 4 7 3 14
If yes, How Frequently % of Total 6.3% 10.9% 4.7% 21.9%
frequently
Count 12 17 5 34
Sometimes
% of Total 18.8% 26.6% 7.8% 53.1%
Count 1 2 1 4
Rarely
% of Total 1.60% 3.10% 1.60% 6.3%
Count 20 35 9 64
Total
% of Total 31.3% 54.7% 14.1% 100.0%

Figure 7.14 Perceived reasons for inability to exploiting arbitrage opportunities

The majority of the respondents (64%) who felt that the arbitrage
opportunities had been difficult to exploit or not exploitable at all were of the
opinion that dearth of liquidity in the options market was the major reason
for their inability to exploiting such opportunities. Higher bid–ask spread
had been cited as the second major reason for this. Notably, higher bid-ask
spreads essentially can be traced to the dearth of liquidity. In operational terms,
172 ● Derivative Markets in India

the investors who feel this as the reason for inability to exploiting arbitrage
indirectly signal liquidity as the main reason.
In sum, therefore, it would be appropriate to conclude that a vast majority
(79%) of such respondents perceived dearth of liquidity as the major reason
for inability to exploiting arbitrage opportunities. This apart, nearly one-tenth
(9%) of such respondents felt that cost associated with such opportunities had
been higher than the profit margins expected from such opportunities.
It is surprising to note that none of the respondents mentioned any of the
market frictions, e.g. short-selling constraint in the market, as a reason for
their inability to exploiting from arbitrage opportunities. The finding suggests
that respondents are either using futures market to overcome the short-selling
constraint or they are not using the arbitrage strategies which require use of
short-sell. However, the latter seems to be the more probable reason as in the
absence of short-selling mechanism, the futures market cannot be expected to
be correctly priced and, therefore, would not be able to identify the arbitrage
in all the cases even if the options market is inefficient. In sum, it would
be reasonable to conclude that the respondents do not seem to be aware of
arbitrage opportunities that require short-selling mechanism in place.

7.4.5 Need of Regulations and Educational Initiatives for the

Betterment of the Options Market
It was also of interest to gauge the opinion of the respondents on aspects
related to regulations and educational initiatives needed for the betterment of
the options market in India. While a sizable majority (70%) of the respondents
felt that a self regulatory mechanism would be preferable to the imposed
regulations, one-fourth of the respondents supported need of regulations
to restore equilibrium in the options market. Another notable finding of
the survey is that the proposal to introduce a price band in order to ensure
the correct pricing of options has been rejected by the majority (53%) of the
respondents. One-sample t-test has been used to statistically validate the
agreement or disagreement on the proposed regulations for the betterment
of the market. The results are summarized in Table 7.8.
The results of the test regarding the regulations as an initiative for the
betterment of the market has been rejected at 5% level of significance. This
validates disagreement of the respondents on the possibility of further
regulations for the development of the market. However, it is surprising to
note that the hypothesis on the proposal of a price band could not be rejected
at 5% level of significance. In other words, while a majority of respondents
rejected the proposal of launching a pricing band for regulating options prices
in the market, the statistical validation of disagreement could not be achieved.
In sum, it would be reasonable to infer that the respondents have a negative
view on regulations for the betterment of the market.
 Table 7.8 One-Sample t-Test for the Statistical Validation of Agreement/Disagreement of Respondents on Regulations for the Betterment of
the Options Market
Test value = 03
95% confidence interval of
Statement Number X s Mean
t df Sig. the difference
difference
Lower Upper
Options prices need to be
64 2.61 1.40 – 2.23 63 0.029 – 0.391 – 0.74 – 0.04
regulated
A price band should be put in
64 3.03 1.32 0.19 63 0.851 0.031 −0.30 0.36
place
Survey Analysis and Findings
●
173
174 ● Derivative Markets in India

Besides, the opinion of respondents was also sought on the need of

educational initiatives for the development of the market. For the purpose, the
survey has attempted to gauge the awareness and sufficiency of educational
initiatives from SEBI, NSE and BSE. Further, the respondents were asked on
need of more education on derivatives to the investors, creating a separate
body for the same and the ownership pattern for such an organization.
The results on level of awareness amongst respondents on existing
educational initiatives of SEBI, NSE and BSE along with their opinion on
the adequacy of such measures have been depicted in Fig. 7.15. The results
demonstrate that a significant segment (74%) of the respondents have been
found to be aware of such initiative; however, one-fourth (26%) of respondents
revealed their ignorance on such educational initiatives. A notable finding of
the survey is that amongst those who were aware of educational initiatives,
a majority (70%) of such respondents felt that the existing initiatives are not
sufficient. In short, the survey indicates that the existing educational initiatives
are not adequate; some more educational initiatives are needed for increasing
participation of investors in the options market.

Figure 7.15 Level of awareness on existing educational initiatives amongst respondents

and their opinion on sufficiency of these initiatives

On the educational initiatives for the investors, the respondents were asked
on three aspects—(i) the need of educating investors more on the derivatives,
(ii) opinion on creating a separate body for the purpose and (iii) the acceptable
ownership pattern for the proposed educational body. The results regarding
the first two aspects have been summarized in Table 7.9 and the responses to
the third aspect have been depicted in Fig. 7.16.
In an overwhelming response to the need of more educational initiatives,
a vast majority (94%) of the respondents felt that enhanced educational
initiatives for investors’ education need to be put in place. As a solution to the
warranted increase in educational initiatives, more than four-fifth (84%) of the
respondents agreed that a separate non-profit organization should be created
to carry out this task. The agreement on these two proposals so observed has
been supported statistically as well (Table 7.9). With respect to the ownership
of the proposed educational organization, a major segment of the respondents
has given first priority to the Public Private Partnership (PPP).
 Table 7.9 One-Sample t-Test for the Statistical Validation of Agreement/Disagreement of Respondents on the Enhanced Educational Initiatives
Test value = 03
95% confidence interval of
Statement Number X s Mean
T df Sig. the difference
difference
Lower Upper
Need of more education on
64 4.55 0.75 16.40 63 0.000 1.547 1.36 1.74
derivatives
Need of a separate body for
64 3.89 0.86 8.31 63 0.000 0.891 0.68 1.10
educating investors on derivatives
Survey Analysis and Findings
●
175
176 ● Derivative Markets in India

Figure 7.16 Respondents’ view point on the ownership pattern of the

proposed educational entity

In short, the respondents of the survey showed a clear agreement on

enhanced education for investors’ on derivatives and strongly felt a need for
a separate educational body with the PPP ownership pattern.

7.5 CONCLUDING OBSERVATIONS

This study attempts to gauge the opinion of trading member organizations on
the state of options pricing in Indian derivatives market. For the purpose, a
survey has been carried out amongst the trading member organizations based
at Delhi and Mumbai. The survey has dealt with five major aspects, namely (i)
level of participation and usage of the options market, (ii) awareness and use
of models for options valuation, (iii) understanding of PCP relationship, (iv)
correctness of options pricing, its impact and existence and exploitability of
arbitrage opportunities and (v) need of regulations and educational initiatives
for the betterment of the market.
With respect to the participation and usage of options market, a vast
majority of participants has been found to be involved in trading in futures
market as nearly 80% of the respondents had less than 20% of their trading
volume relating to derivatives from the options market. The finding clearly
demonstrates that futures are preferred by respondents of the survey; options
carry lower weight in their derivatives portfolio.
It is satisfying to note that the usage of options market in terms of risk
hedging, speculation and arbitrage has been, to a greater extent, in line with
the findings of the L.C. Gupta committee (1998).
In addition, the survey attempted to gauge the level of awareness and use
of valuation models for valuing options. In this respect, it is surprising to note
that only nearly one-fourth (27%) of the respondents have been found to be
actually using the valuation models for the purpose of valuing index options;
nearly one-third (32%) of the respondents have been using such models in
the case of option on individual stock. Another notable finding of the survey
 Survey Analysis and Findings ● 177

is that a major segment of respondents did not understand the concept of

PCP relationship well. It is eloquently corroborated by the fact that nearly
three-fifth (59%) of the respondents was not aware of the relationship. This
might have serious implications for the pricing efficiency of options in India,
as the PCP relationship helps to identify the pricing anomalies/arbitrage
opportunities relating to the relative price of call and put options and, in turn,
helps to restore equilibrium in the market. These findings indicate that the
level of understanding of valuation concepts amongst the market participant
has been considerably low.
Besides, majority of the respondents perceived that options in Indian
securities market are not correctly priced, causing arbitrage opportunities to
exist in the market. However, it is pertinent to note that the dearth of liquidity
has been the major constraint to arbitrageurs to gain from such opportunities.
An equally revealing finding of the survey is that no respondent felt that the
short-selling has been one of the major constraints to exploiting arbitrage in
the market.
As far as the response for the educational initiatives for the betterment of
the market are concerned, the vast majority of respondents strongly feel the
need of enhanced investors’ education on derivatives and recommend for a
separate educational body with the PPP kind of ownership pattern.
In view of the above findings, it would be reasonable to suggest that a
separate educational body should be put in place for educating investors more
on derivatives. Such an organization, in Indian context, could be proposed
in line with its developed counterpart, the USA, where a fully independent
organization known as Options Industry Council (OIC) is devoted to
enhancing investors’ knowledge on options contracts. Such an organization
may be financially supported by the stock exchanges facilitating derivatives
trading in India, e.g. NSE, BSE and MCX along with the market regulator SEBI
and Clearing Corporations in India. Moreover, such an organization should
not only be disseminating education amongst investors, all the employees of
the trading member organization who are actively managing the derivatives
segment should be trained regularly on the derivatives (especially options),
as lack of adequate understanding of the options market amongst participants
could not be ruled out. Such an organization would, to a marked extent,
enhance participation in derivatives market and help the market to operate
closer to the equilibrium, as the liquidity has been the biggest constraint to
the arbitrageurs.
 References

Ackert, L. F., & Tian, Y. S. (1998). The Introduction of Toronto Index

Participation Units and arbitrage opportunities in the Toronto 35 Index
Option Market. Journal of Derivatives, 5(4), 44–53.
Ackert, L. F., & Tian, Y. S. (2000). Evidence on the efficiency of index options
markets. Federal Reserve Bank of Atlanta Economic Review, First Quarter,
40–52.
Ackert, L. F., & Tian, Y. S. (2001). Efficiency in index options markets and
trading in stock baskets, Journal of Banking and Finance, 25, 1607–1634.
Agiakloglou, C., & Newbold, P. (1992). Empirical evidence on Dickey–Fuller-
type tests. Journal of Time Series Analysis, 13, 471–83.
Akgiray, V. (1989). Conditional heteroskedasticity in time series of stock
returns: evidence and forecasts. Journal of Business, 62, 55–80.
Alexander, C. (1998). Risk Management and Analysis: Measuring and Modelling
Financial Risk. John Wiley & Sons, New York.
Alexander, C., & Leigh, C. T. (1997). On the covariance matrices used in value-
at-risk model. Journal of Derivatives, 4(3), 50–62.
Aragonés, J. R., Blanco, C., & Estévez, P. G. (2007). Neural network volatility
forecasts. Intelligent Systems in Accounting, Finance and Management, 15,
107–121.
Baesel, J.B., Shows, G., & Thorpe, E. (1983). The cost of liquidity services in
listed options. Journal of Finance, 38(3), 989–995.
Baillie, R. T., & DeGennaro, R. P. (1990). Stock returns and volatility. Journal
of Financial and Quantitative Analysis, 25, 203–214.
Baillie, R. T., Bollerslev. T., & Mikkelsen, H. O. (1996). Fractionally Integrated
Generalized Autoregressive Conditional Heteroskedasticity. Journal of
Econometrics, 74(1), 3–30.
Beckers, S. (1981). Standard deviations implied in option prices as predictors
of future stock price variability. Journal of Banking and Finance, 5, 363–381.
Berg, E., Brevik, T., & Saettem, M. (1996). An examination of the Oslo Stock
Exchange options market, Applied Financial Economics, 6, 103–113.
Bharadwaj, A. & Wiggins, J.B. (2001). Box Spread and PutCall Parity Tests for
the S&P 500 Index LEAPS Market. The Journal of Derivatives, 8(4), 62-71.
184 ● References

Bhattacharya, M. (1983). Transactions data tests of efficiency of the Chicago

Board Options Exchange. Journal of Financial Economics, 12, 161–185.
Billingsley, R. S., & Chance, D. (1985). Options market efficiency and the box
spread strategy, Financial Review, 20(4), 287–301.
Black, F. (1975). Fact and fantasy in the use of options. Financial Analysts
Journal, 31(4), 61–72.
Black, F. (1976). The Pricing of Commodity Contracts. Journal of Financial
Economics, 3, 167–179.
Black, F., & Scholes, M. (1973). The pricing of options and corporate liabilities.
Journal of Political Economy, 81, 637–659.
Blair, B.J., Poon, S-H and Taylor S.J. (2001). Forecasting S&P100 Volatility:
The Incremental Information Content of Implied Volatilities and High-
Frequency Index Returns. Journal of Econometrics, 105, 5-26.
Blattberg, R. C., & Gonedes, N.J. (1974). A Comparison of Stable and Student
Distributions as Statistical Models for Stock Prices. Journal of Business, 47(2),
244–280.
Bodurtha Jr., J. N., & Courtadon, G. R. (1986). Efficiency tests of the foreign
currency options market. The Journal of Finance, 16(1), 151–162.
Bollerslev, T. (1986). Generalized autoregressive conditional heteroskedasticity.
Journal of Econometrics, 31, 307–327.
Bollerslev, T., Engle, R. F., & Nelson, D. B. (1994). ARCH Models, in Handbook
of Econometrics, Vol. IV, Robert F. Engle and D.L. McFadden, Eds., North
Holland, Amsterdam, pp 2959–3038.
Brookfield D. (1993). Traded options, capital gains and the term structure of
implied volatilities. Applied Financial Economics, 3, 1–13.
Brooks, C. (2005). Introductory Econometrics for Finance. Cambridge University
Press, Cambridge, UK.
Broughton, J. B., Chance, D. M., & Smith, D. M. (1998). Implied standard
deviations and put–call parity relations around primary security offerings.
The Journal of Applied Business Research, 15(1), 1–12.
Brown, R.L., & Easton, S.A. (1992). Empirical evidence on put–call parity
in Australia: A reconciliation and further evidence. Australian Journal of
Management, 17(1), 11–20.
Brunetti, M., & Torricelli, C. (2007). The internal and cross market efficiency
in index option markets: An investigation of the Italian market. Applied
Financial Economics, 17, 25–33.
Busch, T. (2005). A robust LR test for the GARCH model. Economics Letters,
88, 358-364.
Byoun, S., Chuck C. Y. K., & Park, H. Y. (2003). Expectations hypothesis of the
term structure of the volatility: Evidences from foreign currency and stock
index options. Journal of Financial Econometrics, 1, 126–151.
 References ● 185

Campa, J. M., & Chang, P. H. K. (1995). Testing the expectations hypothesis

on the term structure of volatilities in foreign exchange options. Journal of
Finance, 50(2), 529–547.
Canina, L., & Figlewski, S. (1993). The informational content of implied
volatility. Review of Financial Studies, 6, 659–668.
Capelle-Blancard, G., & Chaudhury, M. (2002). Efficiency Tests of the French
Index (CAC 40) Options Market. Working Paper, McGill Finance Research
Centre (MRFC).
Cavallo, L., & Mammola, P. (2000). Empirical tests of efficiency of the Italian
index options market. Journal of Empirical Finance, 7, 173–193.
Chance, D. M. (1988). Boundary condition tests of bid and ask prices of index
call options. The Journal of Financial Research, 11(1), 21–31.
Chiras, D., & Manaster, S. (1978). The information content of option prices and
a test of market efficiency. Journal of Financial Economics, 6, 213–234.
Chou, R. Y. (1988). Volatility persistence and stock valuations: Some empirical
evidence using GARCH. Journal of Applied Econometrics, 3, 279–294.
Christie, A. A. (1982). The stochastic behaviour of common stock variances.
Journal of Financial Economics, 10, 407–432.
Chu, S. H., & Freund, S. (1996). Volatility estimation for stock index options:
A GARCH approach. Quarterly Review of Economics and Finance, 36(4),
431–450.
Claessen, H. and Mittnik, S. (2002). Forecasting stock market volatility and
the informational efficiency of the DAX-index options market. The European
Journal of Finance, 8, 302-21.
Clark, P. (1973). A subordinated stochastic process model with finite variance
for speculative prices. Econometrica, 41, 135–156.
Corrado, C.J., & Su, T. (1997). Implied volatility skews and stock return
skewness and kurtosis implied by stock option prices. European Journal of
Finance, 3(1), 73–85.
Cotner, J. S., & Horrell, J. F. (1989). An analysis of index option pricing. Journal
of Futures Markets, 9(5), 449–459.
Das, S. R., & Sundaram, R. K. (1999). Of smiles and smirks: A term structure
perspective. Journal of Financial and Quantitative Analysis, 34, 211–239.
Day, T.E., & Lewis, C.M. (1988). The behavior of the volatility implicit in the
prices of index options. Journal of Financial Economic, 22, 103–122.
Day, T.E., & Lewis, C.M. (1992). Stock market volatility and the information
content of stock index options. Journal of Econometrics, 52, 267–287.
Derman, E., & Kani, I. (1994). Riding on the smile. Risk, 7, 32–39.
186 ● References

DeRoon, F., Veld, C., & Wei, J. (1995). A Study on the Efficiency of the Market
for Dutch Long Term Call Options. Center for Economic Research, Working
Paper, Tilburg University.
Dickey, D.A., & Fuller, W.A. (1979). Distribution of the estimators for the
autoregressive time series with a unit root. Journal of American Statistical
Association, 74, 427–431.
Dixit, A., Yadav, S.S. & Jain, P.K. (2007) Testing the expectations hypothesis
on the term structure of volatilities implied by index options: A study of
Indian securities market. Journal of Financial Management & Analysis, 20(2),
38–55.
Diz F., & Finucane, T.J. (1993). Do the option markets really overreact? Journal
of Futures Market, 13(3), 299–311.
Draper, P., & Fung, J. K. W. (2002). A study of arbitrage efficiency between
the FTSE-100 index futures and the options contracts. Journal of Futures
Markets, 22, 3–58.
Dupire, B. (1994). Pricing with a smile. Risk, 7, 18–20.
Easton, S. A. (1994). Non-simultaneity and apparent option mispricing in tests
of put-call parity. Australian Journal of Management, 19(1), 47–60.
Enders,W. (2005), Applied Econometric Time Series, Wiley, New Delhi.
Engle, R.F. (1982). Autoregressive conditional heteroskedasticity with estimates
of the variance of the U.K. Econometrica, 50, 987–1008.
Engle, R.F., Hong, T., Kane, A., & Noh, J. (1991). Arbitrage valuation of variance
forecasts. Advanced Futures and Options Research, 6, 393–415.
Engle, R.F., Kane, A., & Noh, J. (1996). Index-option pricing with stochastic
volatility and the value of accurate variance forecasts. Review of Derivatives
Research, 1, 139–157.
Engle, R. F. (1993). GARCH 101: The use of ARCH/ GARCH models in applied
econometrics. Journal of Economic Perspectives, 15(4), 157–168.
Engle, R. F. (2002). Dynamic Conditional Correlation: A Simple Class of
Multivariate Generalized Autoregressive Conditional Heteroskedasticity
Models. Journal of Business and Economic Statistics, 20, 339–350.
Engle, R. F., & Gonzalez-Rivera, G. (1991). Semiparametric ARCH Models.
Journal of Business and Economic Statistics, 9, 345–360.
Engle, Robert F., & Ng, V. (1993). Measuring and Testing the Impact of News
on Volatility. Journal of Finance, 48, 1749–1778.
Engle, R.F. & Patton, A.J. (2001). What Good is a Volatility Model? Quantitative
Finance, 1, 237–245.
Evnine, J. & Rudd, A. (1985). Index Options: The Early Evidence, Journal of
Finance, 40, 743–756.
 References ● 187

Fama, E. F. (1965). The Behaviour of Stock Market Prices. Journal of Business

38(1), 34 –105.
Feinstein, S.P. (1989). A Theoretical and Empirical Investigation of the Black–
Scholes Implied Volatility. Dissertation, Yale University, New Haven, CT.
Figlewski, S. & Webb, G. (1993). Options, Short Sales and Market Completeness.
Journal of Finance, 48(2), 761–777.
Figlewski, S. (1997). Forecasting Volatility. Financial Markets, Institutions &
Instruments, 6, 1–88.
Figlewski, S. (2004). Forecasting Volatility. Working paper, New York University,
Stern School of Business.
Finnerty, J. (1978). The Chicago Board Options Exchange and Market Efficiency.
Journal of Financial and Quantitative Analysis, 13(1), 29–38.
Finucane, T. J. (1991). Put–call parity and expected returns. Journal of Financial
and Quantitative Analysis, 26(4), 445–457.
Fleming, J. (1998). The quality of market volatility forecasts implied by S&P100
index option prices. Journal of Empirical Finance, 5, 317-45.
Frankfurter, G. M., & Leung, W. K. (1991). Further analysis of the put–call
parity implied risk-free interest rate. The Journal of Financial Research 14
(3), 217–232.
Fung, J.K.W. & Chan, K.C. (1994). “On the Arbitrage Free Pricing Relationship
between Index Futures and Index Options: A Note,” Journal of Futures
Markets, 14(8), 957–962.
Fung, A. K. W., & Fung, J. K. W. (1997): Misprising of index futures contracts:
A study of index futures versus index options, The Journal of Derivatives
(Winter), 37–44.
Fung, J. K. W., & Mok, H. M. K. (2001). Index options-futures arbitrage: A
comparative study with bid/ask and transaction data. The Financial Review
36, 71–94.
Fung, J. K. W., Cheng, L. T. W., & Chan, K. C. (1997). The intraday pricing
efficiency of Hang Seng Index options and futures markets. Journal of Futures
Markets, 17, 327–331.
Galai, D. (1977). Tests of market efficiency of the Chicago Board Options
Exchange. Journal of Business, 50(2), 167–197.
Galai, D. (1978). Empirical tests of boundary conditions for CBOE options.
Journal of Financial Economics, 6, 187–211.
Garay, U., Ordonez, M. C., & Gonzalez, M. (2003). Tests of the put–call parity
relation using options on futures on the S&P 500 Index. Derivatives Use,
Trading and Regulation, 9, 259–280.
Gemmill, G. (1986). The Forecasting Performance of Stock Options on the
London Traded Options Market. Journal of Business Finance and Accounting.
13(4), 535-546.
188 ● References

Geweke, J. (1986). Modelling Persistence in Conditional Variances: A Comment.

Econometric Review, 5, 57–61.
Ghysels, E., Andrew H., & Renault, E. (1996). Stochastic Volatility. In Handbook
of Statistics: Statistical Methods in Finance, Vol. 14. G.S. Maddala and C.R.
Rao, (eds). Amsterdam: Elsevier Science, pp. 119–191.
Giot, P. (2003). The Information Content of Implied Volatility in Agricultural
Commodity Markets. Journal of Futures Markets, 23(5), 441–454.
Giot, P. (2005). Implied Volatility Indexes and Daily Value at Risk Models.
The Journal of Derivatives, summer, 54–64.
Giot, P., & Laurent, S. (2007). The information content of implied volatility in
light of the jump/continuous decomposition of realized volatility. Journal
of Futures Markets, 27(4), 337–359.
Glosten, L. R., Jagannathan, R., & Runkle, D.E. (1993). On the Relation between
the Expected Value and the Volatility of the Nominal Excess Return on
Stocks. Journal of Finance, 48, 1779–1801.
Goh, L. Y., & Allen, D. (1984). A Note on Put–call Parity and the Market
Efficiency of the London Traded Options Market. Managerial and Decision
Economics, 5(2), 85–90.
Gokcan, S. (2000). Forecasting Volatility of Emerging Stock Markets: Linear
versus Non-linear GARCH Models. Journal of Forecasting, 19(6), 499–504.
Gould, J.P., & Galai, D. (1974). Transactions costs and the relationship between
put and call prices. Journal of Financial Economics, 1, 105–129.
Gray, S. F. (1989). Put call parity: An extension of boundary conditions.
Australian Journal of Management. 14(2), 151–169.
Gujrati, D.N. (2004), Basic Econometrics, Fourth edition, Tata McGraw-Hill,
New Delhi.
Guo, D. (1996). The predictive power of implied stochastic variance from
currency options. Journal of Futures Market, 16(8), 915–942.
Guo, W. & Su, T. (2006). Option put-call parity relations when the underlying
security pays dividends. International Journal of Business and Economics, 5(3),
225–230.
Halpern, P. J., & Turnbull, S. M. (1985). Empirical Tests of Boundary Conditions
for Toronto Stock Exchange. Options. Journal of Finance, 40(2), 481–500.
Harvey, C. R., & Whaley, R. E. (1992). Market Volatility Prediction and
the Efficiency of the S&P100 Index Option Market. Journal of Financial
Economics, 31, 43–73.
Harvey C.R., & Whaley, R.E. (1991b). S&P 100 index option volatility. Journal
of Finance, 46(4), 1551–1561.
Harvey, Andrew C. (1981). The Econometric Analysis of Time Series. Oxford
University Press.
 References ● 189

Heston, S. (1993). A closed form solution for options with stochastic volatility
with applications to bond and currency options. The Review of Financial
Studies, 6, 327–343.
Heynen R., Kemna, A., & Vorst, T. (1994). Analysis of the term structure of
implied volatilities. Journal of Financial and Quantitative Analysis, 29(1), 31–55.
Hull, J., & White, A. (1987). The pricing of options on assets with stochastic
volatility. Journal of Finance, 42, 281–300.
Jameson, M., & William W. (1992). Market Making in the Options Markets and
the Cost of Discrete Hedge Rebalancing. Journal of Finance, 47(2), 765–779.
Johnson, H., & Shanno, D. (1987). Option pricing when the variance is
changing. Journal of Financial and Quantitative Analysis, 22, 143–51.
Joo, T.H., & Dickinson, J.P. (1993). A test of the efficiency of the European
options exchange. Applied Financial Economics, 3, 175–181.
Kamara, A., & Miller Jr., T. (1995). Daily and intraday tests of European put–call
parity. Journal of Financial and Quantitative Analysis, 30, 519–539.
Klemkosky, R. C., & Resnick, B. G. (1979). Put–call parity and market efficiency.
The Journal of Finance, 34, 1141–1155.
Klemkosky, R. C., & Resnick., B. G. (1980). An Ex-Ante Analysis of Put–Call
Parity. Journal of Financial Economics, 8(4), 363–378.
Lamoureux, C.G., & Lastrapes, W.D. (1993). Forecasting stock-return variance:
toward an understanding of stochastic implied volatilities. The Review of
Financial Studies, 6, 293–326.
Latane, H., & Rendleman, R.J. (1976). Standard deviation of stock price ratios
implied by option premia. Journal of Finance, 31, 369–382.
LeBaron B. (1992). Forecast Improvements Using a Volatility Index. Journal of
Applied Econometrics, 7, 137–149.
Lee, C. I., Gleason, K.C., & Mathur, I. (2000). Efficiency Tests in the French
Derivatives Market, Journal of Banking and Finance, 24, 787–807.
Lee, J. H., & Nayar, N. (1993). A transactions data analysis of arbitrage between
index options and index futures. Journal of Futures Markets, 13, 889–902.
Li, K., & Weinbaum, D. (2000). The Empirical Performance of Alternative
Extreme Value Volatility Estimators. Working Paper. http://www.defaultrisk.
com/pp_quant_06.htm.
Li, S., & Alfay, E. (2005). Evidence on the arbitrage efficiency of SPI index futures
and options markets, Economics, Finance & International Competitiveness.
Discussion Papers 194, Queensland University of Technology.
Li, S. (2006). The arbitrage efficiency of Nikkei 225 options market: A put-call
parity analysis, Monetary and Economics Studies, 24,(2), 33–54.
Ljung, G.M., & Box, P.E. (1978). On a measure of lack of fit in time series
models. Biometrica, 66, 66–72.
190 ● References

Malhotra, N. K. (2005). Marketing Research: An applied orientation. Pearsons

Education, Fourth Edition.
Manaster, S., & Rendleman, R.J. (1982). Option Prices as Predictors of
Equilibrium Stock Prices. Journal of Finance 37, 1043–1057.
Mandelbrot, B. (1963). The Variation of Certain Speculative Prices. Journal of
Business, 36(4), 394–419.
Marchand, P.H., Lindley, J.T., & Followill, R.A. (1994). Further evidences on
parity relationships in options on S&P 500 index futures. The Journal of
Futures Markets, 14(6), 757–772.
Martens, M., & Zein, J. (2002). Predicting Financial Volatility: High Frequency
Time Series Forecasts vis-à-vis Implied Volatility. Working Paper, Erasmus
University.
Mayhew, S. (1995). Calculating Implied Parameters from Option Prices. Working
Paper, Haas School of Business, University of California, Berkeley.
Merton, R. C. (1998). Applications of Option-Pricing Theory: Twenty-Five
Years Later. The American Economic Review 88(3), 323–349.
Merton, R. C. (1973a). Theory of rational option pricing. Bell Journal of Economics
and Management Science, 4, 141–183.
Merton, R. C. (1973b). The Relationship between Put and Call Option Prices:
Comment. Journal of Finance, 28(1), 183–184.
Miller, M. H. (1997). Merton Miller on Derivatives. John Wiley & Sons.
Mittnik, S., & Rieken, S. (2000a). Lower-Boundary violations and market
efficiency: Evidence from German DAX-index options market. The Journal
of Futures Markets, 20(5), 405–424.
Mittnik, S., & Rieken, S. (2000b). Put–call parity and the information efficiency
of the German DAX Index options market. International Review of Financial
Analysis, 9, 259–279.
Mixon, S. (2007). The implied volatility term structure of stock index options.
Journal of Empirical Finance, 14, 333–354.
Nance, D.R., Smith Jr., C.W., & Smithson, C.W. (1993). On the Determinants
of Corporate Hedging. Journal of Finance, 48(1), pp. 267–284.
Nattenberg, S. (1994). Option Volatility and Pricing: Advanced Trading Strategies
and Techniques. McGraw-Hill.
Nelson, D. B. (1991). Conditional Heteroscedasticity in Asset Returns: A New
Approach. Econometrica, 59(2), 347–370.
Nilsson, R. (2008). The value of shorting. Journal of Banking and Finance, 32(5),
880–891.
Nisbet, M. (1992). Put–call parity theory and an empirical test of the efficiency
of the London traded options market. Journal of Banking and Finance, 16,
381–403.
 References ● 191

Ofek, E., Richardson, M., & Whitelaw R. F. (2004). Limited arbitrage and short
sales restrictions: Evidence from the options markets. Journal of Financial
Economics, 74, 305–342.
Pantula, S.G. (1986). Modelling Persistence in Conditional Variances: A
Comment. Econometric Review, 5, 71–74.
Parkinson, M. (1980). The Extreme Value Method for Estimating the Variance
of the Rate of Return. Journal of Business, 53, 61–65.
Patell, J. M., & Wolfson, M. A. (1979) Anticipated information releases reflected
in call option prices. Journal of Accounting and Economics, 1, 117–40.
Penttinen, A. (2001). The Sensitivity of Implied Volatility to Expectations of
Jumps in Volatility: An Explanation for the Illusory Bias in Implied Volatility
as a Forecast. Working Paper, New York University.
Phillips, S., & Smith, C. Jr. (1980). Trading costs for listed options: The
implications for market efficiency. Journal of Financial Economics, 8, 179–201.
Poon, S., & Granger, C. W. J. (2003). Forecasting Volatility in Financial Markets:
A Review. Journal of Economic Literature, 41, 478–539.
Pong, S., Shackleton, M.B., Taylor, S.J. and Xu, X. (2004). Forecasting currency
volatility: A comparison of implied volatilities and AR(FI)MA models.
Journal of Banking and Finance, 28, 2541-63.
Poon, S., & Granger, C. W.J. (2005). Practical Issues in Forecasting Volatility.
Financial Analyst Journal, 61(1), 45–46.
Poterba J.M., & Summers, L.H. (1986). The persistence of volatility and stock
market fluctuations. The American Economic Review, 76(5), 1142–1151.
Poteshman, A. (2001). Underreaction, overreaction, and increasing misreaction
to information in the options market. Journal of Finance, 56, 851–876.
Puttonen, V. (1993). Boundary conditions for index options: Evidence from
the Finnish market. The Journal of Futures Market, 13(5), 545–562.
Report of the L C Gupta Committee on Derivatives, Securities and Exchange Board
of India, March 1998. http://www.sebi.gov.in/commreport/main.html
Rogers, L.C.G., & Satchell, S.E. (1991). Estimating Variance from High, Low
and Closing Prices. Annals of Applied Probability, 1, 504–512.
Rogers, L.C.G., Satchell, S.E., & Yoon, Y. (1994). Estimating the Volatility of
Stock Prices: A Comparison of Methods that Use High and Low Prices.
Applied Financial Economics, 4(3), 241–247.
Ross, S. A. (1976). Options and efficiency. Quarterly Journal of Economics 90,
75–89.
Rubinstein, M. (1994). Implied binomial trees. Journal of Finance, 49, 771–818.
Schmalensee, R. & Trippi, R.R. (1978). Common Stock Volatility Expectations
Implied by Option Premia. Journal of Finance, 33, pp. 129–147.
Schwert, G.W. (1989). Why does stock market volatility change over time?
Journal of Finance, 44, 1115–54.
192 ● References

Schwert, G.W., & Seguin, P.J. (1990). Heteroskedasticity in stock returns. Journal
of Finance, 45, 1129–1155.
Scott, E., & Tucker, A.L (1989). Predicting currency return volatility. Journal
of Banking and Finance, 13, 838–851.
Scott, L. O. (1987) Option pricing when the variance changes randomly:
theory, estimation and an application. Journal of Financial and Quantitative
Analysis, 22, 429–38.
Sentana, E. (1995). Quadratic ARCH Models. Review of Economic Studies, 2,
639–661.
Shastri, K., & Tandon, K. (1985). Arbitrage Tests of the Efficiency of the Foreign
currency Options Market. Journal of International Money and Finance, 4,
455–468.
Shastri, K., & Tandon, K. (1986). An empirical test of a valuation model for
American options on futures contracts. Journal of Financial and Quantitative
Analysis, 10, 377–392.
Skinner, D. J. (1989). Options Markets and Stock Return Volatility. Journal of
Financial Economics, 23, 61–78.
Smith Jr., C. W., (1976), “Option Pricing: A Review,” Journal of Financial
Economics, 3(1&2), 3–51.
Smith, C.W., & Stulz, R.M. (1985). The Determinants of Firm’s Hedging Policies.
Journal of Financial and Quantitative Analysis, 20(4), 391–405.
Srivastava, S., Yadav, S. S., & Jain, P.K. (2008). Derivatives Trading in Indian
Stock Market: Broker’s Perception. IIMB Management Review, 20(3), 311–323.
Stein J. (1989). Overreactions in the options market. Journal of Finance, 44(4),
1011–1023.
Sternberg, J. S. (1994). A Reexamination of Put–call parity on index futures.
The Journal of Futures Markets, 14(1), 79–101.
Steven, L. (2006). The Arbitrage Efficiency of the Nikkei 225 Options Market:
A Put–Call Parity Analysis, Monetary and Economic Studies, 24(2), 33–54.
Stoll, H. R. (1969). The relationship between put and call option prices. The
Journal of Finance, 24(5), 801–824.
Stoll, H. R. (1989). Inferring the Components of the Bid–Ask Spread: Theory
and empirical Tests, Journal of Finance, 44(1), 115–134.
Stulz, R. M. (2004). Should We fear Derivatives? Working Paper No. 10574.
National Bureau of Economic Research, Cambridge.
Takezaba N., & Shiraishi, N. (1998). A note on the term structure of implied
volatilities for the Yen/ U.S. dollar currency options. Asia-Pacific Financial
Markets, 5, 227–236.
Taylor, S. L. (1990). Put–call parity: Evidence from the Australian options
market. Australian Journal of Management, 15(1), 203–216.
 References ● 193

Taylor, S. L. (1994). Modeling stochastic volatility: A review and comparative

study. Mathematical Finance, 4, 183–204.
Trennepohl, G. L., & Dukes, W. P. (1979). CBOE Options and Stock Volatility.
Review of Business and Economic Research 14(3), 49–60.
Trippi, R. (1977). A test of option market efficiency using a random-walk
valuation model. Journal of Economics and Business, 29, 93–98.
Tufano, P. (1995). Using Derivatives: What Senior managers must know.
Harvard Business Review, January–February, 35–36.
Turvey, C. (1990). Alternative Estimates of Weighted Implied Volatilities from
Soybean and Live Cattle Options. Journal of Futures Market, 10(4), 353–366.
Varma, J. R. (2002). Mispricing of Volatility in Indian Index Options Market.
Working Paper. Indian Institute of Management, Ahmedabad.
Vasilellis, G. A., & Meade, M. (1996). Forecasting Volatility for Portfolio
Selection. Journal of Business Finance and Accounting, 23, 125–143.
Vijh A. (1990). Liquidity of the CBOE Equity Options. Journal of Finance, 45(4),
1157–1179.
Vipul. (2008). Cross-market efficiency in the Indian derivatives market: A test
of put–call parity. The Journal of Futures Markets. 28(9), 889–910.
Wagner, D., Ellis, D.M., & Dubofsky, D.A. (1996). The factors behind put
call parity violations of S&P 100 index options. The Financial Review, 31(3),
535–552.
West, K.D., Edison, D.H.J., & Cho, D. (1993). A utility based comparison of
some models of exchange rate volatility. Journal of International Economics,
35, 23–45.
Whaley, R. E. (1982). Valuation of American Call Options on Dividend Paying
Stocks: Empirical Tests. Journal of Financial Economics, 10, 29–58.
Whaley, R. E. (1993). Derivatives on Market Volatility: Hedging Tools Long
Overdue. Journal of Derivatives 1, 71–84.
Wilson, W. W., & Fung, H. (1994). Put–call parity and arbitrage bounds for options
on grain futures. American Journal of Agricultural Economics, 73(1), 55–65.
Xu, X., & Taylor, S.J. (1994). The term structure of volatilities implied by
foreign exchange options. Journal of Financial and Quantitative Analysis, 29
(1), 57–74.
Xu, X., & Taylor, S.J. (1995). Conditional Volatility and the Informational
Efficiency of the PHLX Currency Options Market. Journal of Banking and
Finance, 19, 803–821.
Zakoian, J.M. (1991). Threshold Heteroscedastic Models. Technical Report,
INSEE.
Zhang, Z., & Lai, R. N. (2006). Pricing efficiency and arbitrage: Hong Kong
derivatives markets revisited. Applied Financial Economics, 2006, 16, 1185–
1198.
 Glossary

American option An option contract that can be exercised on or up to

the expiration date of the contract.
AR Autoregressive models are those models that include
lag(s) of the dependent variable as explanatory
variables for modeling the dependent variable.
ARCH Autoregressive Conditional Heteroscedasticity—A
time-series model used to deal with the financial time-
series data exhibiting variance that is not constant over
time. These models use past observations of variance
to forecast future variance.
ARIMA Autoregressive Integrated Moving Average models are
used to model the time-series data that exhibit a trend
along with the ‘autoregressive’ and ‘moving average’
characteristics.
ACF Autocorrelation Function represents the correlation
sequence of a random time series with itself. This is
used to test the autoregressive and moving average
tendency of a time series data.
Bid The price at which market participants are willing to
buy an asset.
Brokerage firm A firm or person engaged in executing orders to buy
or sell futures contracts for customers. A full service
broker offers market information and advice to assist
the customer in trading. A discount broker simply
executes orders for customers.
Call option An option contract that gives a right, but not an
obligation, to its holder to buy the underlying asset at
a specified price.
Cash price Also called spot price and represents the current market
price of an asset.
Cash settlement Final disposition of open positions on the last trading
day of a contract month. Such settlement occurs in
the market where there is no physical delivery or for
180 ● Glossary

the assets where physical delivery is not possible (e.g.

index).
Clearing house A market intermediary associated with the stock
exchange, responsible for the settlement of trading
accounts, clearing trades, collecting, regulating
delivery and reporting trading data.
Conditional models Time-series techniques with explicitly specified
dependence on the past sequence of observations.
EGARCH Exponential Generalized Autoregressive Conditional
Heteroscedasticity is one of the time-series models
used to estimate the variance of a time series where
innovations contribute asymmetrically, depending on
their sign, to the variance of the series.
European options Options that can be exercised only at the expiration
date of the contract.
Exercise price (also The price at which the holder (buyer) can purchase
known as strike price) or sell the underlying asset.
Expiration date The date beyond which the contract has no value.
Forward contract A private agreement between buyer and seller for the
future delivery of an asset at an agreed price.
Futures contract A standardized form of forward contract. The futures,
unlike forwards, are traded on an exchange. These
contracts typically include mark-to-market feature to
contain the risk that emanates from daily movement
in the value of the underlying asset.
GARCH Generalized Autoregressive Conditional Hetero-
scedasticity is a time-series model used to estimate the
variance of a time series where innovations contribute
symmetrically, irrespective of their sign, to the variance
of the series.
Hedging The process of protecting the value of an asset against
the unfavourable movement in the market. This
typically aims at reducing variations in the value of
an asset.
Heteroscedasticity It refers to the time-dependent variance of time-series
data.
Homoscedasticity It refers to the time-independent nature of volatility
(constant volatility over time) of time-series data.
Long position When a market participant purchases an asset or a
contract to purchase the underlying asset, he is said
to have taken a long position.
 Glossary ● 181

MA model Moving average models include past observations

of the innovations (noise) in the forecast of future
observations of the dependent variable of interest.
Mark-to-market The daily adjustment of a futures account in response
to changes in the price of a futures contact to reflect
profits and losses. This represents the risk containment
measures in the futures market.
MSE Mean square of error–A statistical measure used to
examine the forecasting ability of different models
under consideration by looking at the deviation of the
model generated forecasts from the realised values.
Non-simultaneity This refers to the phenomenon of different timings of
closing transactions in the two markets (e.g. the options
market and the underlying’s cash market).
Open interest Total number of futures or options that have not yet
been settled.
Option contract A contract which gives a right, but not an obligation,
to its holder to buy or sell the underlying asset at a
specified price within a specified time period.
Option buyer One who purchases an option and pays a premium.
Option seller or One who writes an option and receives a premium.
Option writer
Put option An option contract which gives a right, but not an
obligation, to its holder to sell the underlying asset at
a specified price.
Settlement price A figure determined by the closing range of prices
after a trading session that is used to calculate gains
and losses in futures market accounts.
Short position This refers to selling of an asset without actually having
it. The process of short-selling is generally executed
through a borrowing and lending mechanism, which
allows this process to take place in a legal way.
Speculator A market participant who attempts to make profit from
the anticipated changes in the price of an asset.
Trading volume The number of contracts traded during a specified
period of time, e.g. daily.
Volatility An annualized measure of the fluctuation in the price
of an asset.
 Authors’ Profiles
Alok Dixit is currently working as an Assistant Professor
of Finance & Accounting at Indian Institute of Management
Lucknow (IIM-Lucknow), India. He teaches Management
Accounting, Investment Management and Quantitative
Applications in Finance. He obtained his doctoral degree
(Ph.D.) from the Department of Management Studies,
Indian Institute of Technology Delhi (IIT-Delhi) on
the topic, “Pricing Efficiency of S&P CNX Nifty Index
Options”. He is a recipient of Junior Research Fellowship
of the University Grants Commission (UGC-JRF) in
management. He has published research papers in the journals of national and
international repute. E-mail: [email protected]
Surendra S Yadav is currently Professor of Finance at the
Department of Management Studies, Indian Institute of
Technology Delhi (IIT-Delhi), India. He teaches Corporate
Finance, International Finance, International Business,
and Security Analysis and Portfolio Management. He
has been a visiting professor at the University of Paris,
Paris School of Management, INSEEC Paris, and the
University of Tampa, USA. He has published nine books
and contributed more than 130 papers to research journals
and conferences. He has also contributed more than 30
papers to financial/economic newspapers. E-mail: [email protected]
P K Jain is Professor of Finance & Modi Chair Professor
at the Department of Management Studies, Indian
Institute of Technology Delhi (IIT-Delhi), India. He
has been Dalmia Chair Professor as well. Recently, he
has been awarded with ‘Best Faculty Award’ at IIT
Delhi. He has nearly 40 years of teaching experience in
subjects related to Management Accounting, Financial
Management, Financial Analysis, Cost Analysis and
Cost Control. He has been a visiting faculty at AIT
Bangkok, Howe School of Technology Management
at Stevens Institute of Technology, New Jersey; and ICPE, Ljubljana. He has
authored three well-known text books published by TMH and more than 10
research monographs. He has contributed more than 140 research papers in
journals of national and international repute. E-mail: [email protected]