Introduction to Market efficiency

It was generally believed that securities market were extremely efficient in reflecting information
about individual stocks and about stock market as a whole. The accepted view that when
information arises; the news spreads very quickly and is incorporated into the prices of the
securities without delay. Thus, neither technical analysis, which is the study of past stock prices
in an attempt to predict future prices, nor even fundamental analysis, which is the analysis of
financial information such as companys earnings, asset value etc. to help the investors select
undervalued stocks, would enable an investor to achieve returns greater than those that can be
obtained by holding a randomly selected portfolio stocks with comparable risk.
The efficient market hypothesis is associated with the ideas of a random walk which says that
if the flow of information is unimpeded and information is immediately reflected in the stock
prices, then tomorrows price change will reflect only tomorrows news and will be independent
of the price changes today. But then news is by definition unpredictable and thus, resulting price
changes must be unpredictable and random. As a result, price fully reflect all known information
and even uninformed investors buying a diversified portfolio at the tableau of prices given by the
market will obtain a rate of return generous as that achieved by the experts.
If prices wander randomly, then this poses a major challenge to market analysts who try to
predict the future path of security prices. Informational efficiency of the market takes three forms
depending upon the information reflected by security prices. First, EMH in its weak form states
that all information impounded in the past price of a stock is fully reflected in the current price of
the stock. Therefore, information about recent or past trend in stock prices is of no use in
forecasting the future prices. Clearly it rules out the use of technical analysis in predicting the
future prices of securities. The semi strong form takes the information set one step further and
includes all publicly available information. There are number of information available to
investors which are of their potential interest. So analyzing annual reports and other published
data with a view to make profit in excess is not possible because market prices ha already
adjusted to any good or bad news contained in such reports as soon as they were revealed. The
EMH in its strong form states that current market price reflect all both public and private
information and even insiders would find it impossible to earn abnormal returns in stock market.

Stock market seasonalities:
EMH proposes that it is not possible to outperform the market through market timing and stock
selection. However in the context of financial markets and particularly in the context of equity
market seasonal component have been recorded. They are called calendar anomalies. The
presence of seasonalities in stock returns violates the weak form of market efficiency because
equity prices are no longer random and can be predicted based on past pattern. For instance if
there is an evidence proving day of the week effect, investors may adopt a trading strategy of
selling securities on Fridays an buying on Mondays in order to make abnormal profits. One of
the explanations put forward for the existence of seasonality in stock returns is the tax-loss-
selling hypothesis. In USA December is tax month. Thus, the financial houses sell shares whose
values have fallen to book losses to reduce their tax, resulting in the decline of the stock prices.
However as soon as December ends, people start acquiring shares and as a result stock price
bounce back. This leads to higher returns in the month of January. This is called January effect.
The academicians and practitioners have documented many research works on the seasonality
and associated behavior of securities market all over the world. Among others the most widely
mentioned seasonal effects and market anomalies are January effect, Monday or weekend effect,
holiday effect and small firm effect, to mention a few.











Weekend effect or Monday effect
Mondays average return is significantly lower than the other days average returns. It refers to
the tendency of stocks to exhibit relatively large returns on Fridays compared to those on
Mondays. Some theories explains the effect attribute the tendency for companies to release bad
news on Fridays after the market close to depressed stock prices on Mondays. Others state that
the weekend effect might be linked to short-selling, which would affect stock with high short
interest positions. Alternatively, the effect could simply be a result of traders fading optimism
between Friday and Monday. The weekend effect has been regular feature of stock trading
patterns for many years. For examples according to a study by the Federal Reserve, prior to 1987
there was statistically significant negative returns over the weekends. However, the study I
mention that this negative returns had disappeared in the period from post 1987 to 1998. Since
1998, volatility over the weekends has increased again, and the phenomenon of the weekend
effect remains a much debated topic.
Literature review:
French (1980) was one of the first to document the weekend effect. He found Monday returns to
standard & poors 500 index were systematically negative. Gbbons and Hess (1981) confirmed
the thrust of Frenchs findings using the CRSP equal-weighted and value-weighted market return
indices. Since these two papers were published, there have been a number of additional papers
examining this apparent anomaly.
Lakonishok & Maberly (1990) attribute some of the Monday-Friday differential returnsto the
differential trading patterns of institutions and individuals. Damodaran (1989) exploreswhether a
tendency of corporations to release bad news on Friday after the market close could account for
depressed Monday share price; he reports evidence of only a weak connection.
Chaudhuri (1991) supported the presence of weekend effect in the daily return of BSE sensitive
index for the period June 1988- January 1990 through Kruskal-Wallis Test. The day of the week
effect in Indian market has been examined by Roger Ignatius (1992) and Golak Nath and Manoj
Dhalvi (2004) for the periods of 1979-1990 and 1999-2003 respectively. Both of them confirmed
the existence of a weak form of weekend effect. However, none of them could give a concrete
reason to explain the weekend effect in India.
Ho Richard and Cheung (1994) studied the seasonal pattern in volatility of Asian stock market.
Using Levene (1960) test, they report that there exist day-of-the- week variations in volatility in
most of the emerging Asian markets.
In Indian market there are a few studies which support the weekend effect. Sunil Poshakwale
(1996) examined the day-of-week effect on BSE using daily BSE national index data for the
period 1987-1994. By applying first order auto correlation test, the study confirmed the weekend
effect as the returns achieved on Fridays are significantly higher compared to the rest of the days
of the week. By employing multiple regression models, Arumugam (1998-99) tested the week
end effect by using BSE sensitive index during April 1979-March 1997. He observed positive
Friday return and significant negative Monday return in bear phase.
Anshuman & Goswami (1999) investigated the weekend effect by using equally weighted
portfolio constructed from 70 stocks listed on the BSE during the period April 1991-march 1996.
Based on empirical investigation, the results evidenced above average positive Friday returns
and below average negative Tuesday return during that period. S. Amanulla & M Tthiripalraju
(2001) using dummy variable regression studied the weekend effect using 82 individual stocks
traded in BSE during the period January 1990-december 1999. Besides this the study was
conducted using three stock market price indices i:e BSE sensitive index, S&P CNX Nifty index.
The study documented that traditional weekend effect was generally observed only in the period
of ban on carry forward transactions during March 11, 1994-January 15, 1996. The revised
modified carry forward transactions have supported the existence of the stock return variations
by showing positive Wednesday return and negative Tuesday during that period and hence
produce a day of the week effect. It also documents a reversal in week-end effect (i:e positive
Monday return and negative Friday return) in these periods.
Dr. Rengasamy Elango and Nabila Al Macki (2007) investigated Monday effect in the major
indices of the NSE i:e S&P CNX Nifty, S&P CNX Defty and CNX Nifty junior for a period
from 1999-2007. Non-parametric and parametric tests like Kruskal Wallis test, Mann Whitney U
test and dummy variable regression model was used. The study evidenced mixed results
indicating that the Monday returns are negative and low in case of two or three indices. And
surprisingly Wednesday have yielded the highest mean returns across indices.
Deepa Mangala (2008) tested the presence of day-of-the-week effect in S&P CNX Nifty index
over a period of seventeen years period commencing from January 1991 through December
2007. K-W test, Mann Whitney U test was adopted for the study. The day of the week effect is
evident as the mean return achieved on Wednesday is significantly higher as compared to rest of
the days of the week. The mean return is most negative on Tuesday.
A recent study was conducted by Ankur singhal and vikram bahure (2009) on weekend effect of
stock returns in Indian market taking opening and closing prices of three major operational
indices in india i:e BSE sensex, BSE 200 and S&P Nifty. Daily returns were calculated for a
period from april 2003 april 2008 and the study was conducted using dummy variable regression
indicating lower returns on Mondays and maximum returns on Fridays across different indices.
Ash Narayan Sah (2009) studied on S&P Nifty from 1997 to 2009 using dummy variable
regression model. The evidence did not find the presence of weekend effect.
Very recent analysis on Monday effect was done by P. Nageswari, M. Selvam and J Gayathri
(2011), taking S&P Nifty and S&P CNX 500 index and using the dummy variable regression
model for a period of April 2002-March 2010. The result of the study found that there was
highest mean return earned on Friday and the lowest mean return earned on Monday for sample
indices. The seasonalities results indicate that there were no significant days of the week effect.


















Objective of the Study:
a) To examine the days of the week effect in the returns of S&P CNX NIFTY.
b) To examine the weekend effect in S&P CNX NIFTY returns
c) To examine the days of the week effect in the returns of CNX NIFTY junior.
d) To examine the weekend effect in CNX NIFTY junior returns.


RESEARCH METHODOLOGY

The study has been conducted on the National Stock Exchange (NSE) is the16th largest stock
exchange in the world by market capitalization and largest in India by daily turnover and number
of trades, for both equities and derivative trading. NSE has a market capitalization of
around US$985 billion and over 1,646 listings as of December 2011. Though a number of other
exchanges exist, NSE and the Bombay Stock Exchange are the two most significant stock
exchanges in India and between them are responsible for the vast majority of share transactions.
The NSE's key index is the S&P CNX Nifty, known as the NSE NIFTY (National Stock
Exchange Fifty), an index of fifty major stocks weighted by market capitalisation.
NSE is mutually owned by a set of leading financial institutions, banks, insurance companies and
other financial intermediaries in India but its ownership and management operate as separate
entities. There are at least 2 foreign investors NYSE Euronext and Goldman Sachs who have
taken a stake in the NSE. It is the second fastest growing stock exchange in the world with a
recorded growth of 16.6%.
Indices Studied Upon
The present considers the daily indices reported by NSE. S&P CNX Nifty and CNX Nifty Junior
were chosen as the indices to be studied upon. As stated, the study attempts to examine the
presence of day-of-the-week effect in Indias premier stock exchange, hence the selection of the
indices for the study was based on certain logical considerations.


S&P CNX Nifty
The S&P CNX Nifty is a well diversified 50 stock index accounting for 22 sectors of the
economy. It is used for a variety of purposes such as benchmarking fund portfolios, index based
derivatives and index funds. S&P CNX Nifty is owned and managed by India Index Services and
Products Ltd. (IISL), which is a joint venture between NSE and CRISIL. IISL is India's first
specialised company focused upon the index as a core product. IISL has Marketing and licensing
agreement with Standard & Poor's (S&P), who world leaders are in index services.
The S&P CNX Nifty Index represents about 66.90% of the free float market
capitalization of the stocks listed on NSE as on December 30, 2011.
The total traded value for the last six months of all index constituents is approximately
56.58% of the traded value of all stocks on the NSE.
Impact cost of the S&P CNX Nifty for a portfolio size of Rs.50 lakhs is 0.08%.
S&P CNX Nifty is professionally maintained and is ideal for derivatives trading
CNX Nifty Junior
The next rung of liquid securities after S&P CNX Nifty is the CNX Nifty Junior. It may be useful to think of
the S&P CNX Nifty and the CNX Nifty Junior as making up the 100 most liquid stocks in India. As with
the S&P CNX Nifty, stocks in the CNX Nifty Junior are filtered for liquidity, so they are the most liquid of
the stocks excluded from the S&P CNX Nifty. The maintenance of the S&P CNX Nifty and the CNX Nifty
Junior are synchronized so that the two indices will always be disjoint sets; i.e. a stock will never appear in
both indices at the same time. Hence it is always meaningful to pool the S&P CNX Nifty and the CNX
Nifty Junior into a composite 100 stock index or portfolio.

The CNX Nifty Junior Index represents about 11.68% of the free float market
capitalization of the stocks listed on NSE as on December 30, 2011.
The total traded value for the last six months of all index constituents is approximately
13.73 % of the traded value of all stocks on NSE.
Impact cost for CNX Nifty Junior for a portfolio size of Rs.25 lakhs is 0.13%.

Collection of Data
The data comprise daily closing prices of the national stock exchange (NSE) from 1.1.1998 to
31.12.2009 covering a period of about eleven years. The required data has been downloaded
from NSE website (www.nseindia.com).
The indices included for the study along with the period covered are given below
Sl. No. Index Period No. of Observations
1. S&P CNX NIFTY 1.1.1998 to 31.12.2009 3000
2. CNX NIFTY JUNIOR 1.1.1998 to 31.12.2009 3000

The Indian stock exchanges, opens on Monday and close on Fridays. For this present study we
have taken into consideration all the working days. All the data points where returns are zero
have been eliminated. The various hypothesis tested have been listed below. Both the non-
parametric and parametric tests have been adopted for the study.
Methodology for Data Calculation
1. Computation of daily returns
The daily returns on NSE index were computed using the first differences of the logarithmic
price index. This approach of logarithmic transformation of the time series data was first
suggested by Osborne (1959).
The returns are calculated as:
Rt = [ln (Pt/Pt-1)]*100
The lognormal returns follow the normal distribution more closely than returns. (Lauterbach and
Ungar (1995).
Where, Rt is the daily return from the index
P is the price index,
P
t
represents the current closing index price of the day.
P
(t-1)
represents the immediate preceding index price.
In order to examine the presence of the day-of-the week effect, the following null hypothesis has
been tested:
Hypothesis (Ho): a
1
=a
2
=a
3
= a
4
=a
5
Here a
1
, a
2
represents mean returns of different trading days of the week. The null hypothesis
implies that there is no significant difference in mean returns across the trading days.
Hypothesis (Ho): Each a
i
is tested for significance (difference from zero)
If this hypothesis is rejected, it would imply that the mean daily returns a
i
is significantly
different from each other, i.e. there is seasonality in returns across different days of the week.

2. Descriptive Statistics

Parametric tests like mean, standard deviation, skewness and kurtosis have been applied to
study the distribution pattern of the daily returns across the week.

3. Tests for seasonality

Wilcoxon Mann-Whitney (U) test:
Non-parametric methods have been employed to test the seasonality because of their
robustness arising from lack of restrictive assumptions such as population normality and
homoscedastic variance. Wilcoxon Mann-Whitney (U) pair-wise test has been applied to
capture the Monday effect. This test examines if the average Monday return is different and
statistically significant from returns generated from each of the remaining four days of the
week, based on ranking differences in pair-wise observations
The Mann-Whitney "U" is then given by:

n
1
(n+1)
U
1
=R
1
-
2
Where n
1
is the two sample size for sample 1, and R1 is the sum of the ranks in sample 1.
Analysis of Variance
Analysis of variance (ANOVA) model is that model where the dependent variable is
quantitative in nature and all the independent variables are categorical in nature. This method is
adopted to find out whether the daily returns are dependent on the trading days or independent of
the days of the week
Dummy variable regression model
To examine the weekend effect and days of the week effect, the following dummy variable
regression model is specified as follows:
Model: R
t
=
1
d
1
+
2
d
2
+
3
d
3
+
4
d
4
+
5
d
5
+ e
t
Where:

R
t
is the return on day t,
I(t) refers to index price on day t;
a1 to a5 are the mean return for each day-of-the-week;
d
1
=dummy variable equal to 1 if t is a Monday otherwise 0.
d2= dummy variable equal to 1 if t is a Tuesday otherwise 0.
d3= dummy variable equal to 1 if t is a Wednesday otherwise 0.
d4= dummy variable equal to 1 if t is a Thursday otherwise 0.
d5= dummy variable equal to 1 if t is a Friday otherwise 0.
e
t
is the random error term for day t.
The intercepts, 1..5, represent the average deviation of each day fro Monday return. Thus, if the
daily returns are equal, one expects the dummy variable coefficients to be statistically close to zero. So,
the coefficients of the regression are the mean returns obtained from Monday to Friday, applying the
ordinary least square.










DESCRIPTIVE STATISTICS AND ANALYSIS
Table 1:

Table 1 above presents the descriptive statistics of the day-of-the-week returns of the two selected indices
CNX NIFTY and CNX NIFTY JUNIOR. Since the study period comprises of 3000 working days of the
exchange, 3000 daily returns for each of the indices are available for analysis. In 12 years of the research
data there are 18 special trading days, 16 Saturdays and 2 Sundays.
CNX NIFTY: A comparison of mean returns across the various trading days of the week makes it clear
that Wednesday account for the highest mean daily return of .269644. The first two trading days of the
week have reported negative returns. The Mondays mean return is the lowest in this index (-.002195).
The standard deviation is highest on Mondays (2.1471026) and lowest on Tuesdays (1.5879098). Hence
the market was more volatile on Mondays. The distribution is a negatively skewed one. The distribution

DAY
Value S&P CNX NIFTY CNX NIFTY JUNIOR
Monday Obsv.
Mean
St.dev.
Skewness
Kurtosis
596
-.002195
2.1471026
-.021
8.384
607
-.021103
2.4213220
-.778
4.628
Tuesday Obsv.
Mean
St.dev.
Skewness
Kurtosis
597
-.021569
1.5879098
-.160
3.539
607
-.005283
1.9523485
-.272
3.197
Wednesday Obsv.
Mean
St.dev.
Skewness
Kurtosis
599
.269644
1.6673894
.103
1.684
599
.383311
1.9874499
-.020
2.007
Thursday Obsv.
Mean
St.dev.
Skewness
Kurtosis
603
.017798
1.6082728
-.199
1.573
601
.014354
1.8472597
-.474
3.217
Friday Obsv.
Mean
St.dev.
Skewness
Kurtosis
586
.031517
1.8409338
-.668
6.159
588
-.030855
2.1160764
-.870
4.145
of daily returns tends to be leptokurtic. The reason for non-normality in the indices could be the high
kurtosis.
CNX NIFTY JUNIOR: This index also reports the highest mean returns on Wednesday
(.383311), Monday, Tuesday and Friday reports negative returns. Tuesday records the lowest return in
this index (.005283). Mondays are the most volatile as the standard deviation on Monday is the highest as
in CNX NIFTY. The distribution of daily return is negatively skewed and tends to be leptokurtic.
Interpretation of the above Analysis:
From the above analysis it may be inferred that Monday effect may not be present in the market and even
the Mondays return have been reported to be the most volatile but there seems to be the presence of day-
of-the-week effect since Wednesdays mean returns have consistently shown higher returns on both the
indices
Table 2: Results of Wilcoxon Mann - Whitney Test

Return Pair
S & P CNX Nifty

S & P CNX Nifty Jr.

Tuesday - Monday
Z -.588 Z -1.334
P Value .556 P Value .182
Wednesday - Monday
Z -2.038 Z -2.659
P Value .042 P Value .008
Thursday - Monday
Z -.780 Z -.612
P Value .436 P Value .541
Friday - Monday
Z -.062 Z -.774
P Value .951 P Value .439
Wednesday - Tuesday
Z -3.046 Z -4.084
P Value .002 P Value .000
Thursday - Tuesday
Z -.793 Z -.696
P Value .428 P Value .486
Friday - Tuesday
Z -.861 Z -.355
P Value .389 P Value .722
Thursday - Wednesday
Z -2.284 Z -2.921
P Value .022 P Value .003
Friday - Wednesday
Z -2.000 Z -3.343
P Value .045 P Value .001
Friday - Thursday
Z -.123 Z -.217
P Value .902 P Value .828


Analysis of U-Test
The size of the P-value for a coefficient says nothing about the size of the effect that variable is
having on the dependent variable i:e on the return.
The higher the p-value, the less we can believe that the observed relation between variables in
the sample is a reliable indicator of the relation between the respective variables in the
population. The higher the percentage of P value, higher will be the chances of no relation
between the independent variable and the return. A p-value of .05 (i.e.,1/20) indicates that there is
a 5% probability that there will be no relation between the variables means there will be 95%
chance that variables are having relation.
The findings of the above table 2 show that every p-value is positive for both the CNX nifty and
CNX nifty junior. In statistical significance testing, the p-value is the probability of obtaining
a test statistic at least as extreme as the one that was actually observed, assuming that the null
hypothesis is true. One often "rejects the null hypothesis" when the p-value is less than 0.05 or
0.01, corresponding respectively to a 5% or 1% chance of rejecting the null hypothesis when it is
true. Results that are significant at the p .01 level are commonly considered statistically
significant. Here in the above result for S&P CNX NIFTY, percentage of all the p-values are
higher than 1 percent except Wednesday- Tuesday return pair, i:e except Wednesday- Tuesday
return pair every other pairs are showing that there are no size of the effect that independent
variables are having on the dependent variable i:e on the return. Hence this concludes that null
hypothesis cant be rejected. Whereas in case of S&P CNX NIFTY JUNIOR, Four pairs are
having less than 1 percent (p .01) and rest of the pairs are above 1 percent level showing no
relation among the variables. One interesting factor is in both the index Friday-Thursday return
pair is showing the highest P value. Friday-Thursday return pair is strongly supporting that there
is no existence of relation among the variables by having 90.2 and 82.8 percentage level p
values.
All the Z values are negative for both the index, just the reverse of P values. ROGER MUNDRY
& JULIA FISCHER have done a U test using Z values. They have found that critical value of z
is 1.96 and if the value of Z exceeds the critical value then we can say that null hypothesis is
rejected.
Our result shows that every Z value is way below the critical value 1.96 at standard significance
level of 5%.Though in our result all the significance level are different but we can safely say that
the z value will not go beyond 1.96 showing that our null hypothesis is accepted and there are no
dependency among variables.
The above test one can safely conclude that there are no fluctuation when it comes to Monday
return. So Weekend effect which was there in the mid 80s as explained by the researchers like
Cross (1973), Gibbons & Hess (1981), Keim & Stambaugh (1984), Theobald and Price (1984),
Jaffe & Westerfield (1985), Harris (1986), Simrlock & Starts (1986), Board and Sutcliffe (1988)
has no existence in todays market making the market efficient.




















Table 3:
The results of Analysis of variance:
Table 2: Regression Model (ANOVA)
Model
Sum of
Squares df Mean Square F Sig.
1 Regression 30.944 5 6.189 .453 .811
a

Residual 7929.744 580 13.672

Total 7960.687 585

R Value 0.062
R
square: 0.004

The table shows that the calculated value of F test statistic (.453) is less than the critical value; hence the
null hypothesis is accepted. This infers that the daily returns of the indices are independent of the trading
days evidencing the absence of weekend effect.


Table 4 Results of dummy variable Regression



Table 4 shows that all the value of the coefficient are positive for S&P CNX NIFTY except Wednesday
indicating that there are fluctuations on return when it comes to Wednesday return, while others have a
positive impact on the dependant variable. In the case of CNX NIFTY JUNIOR all the coefficient values
are positive except Monday and Wednesday showing negative impact on the dependant variable.
Day Parameter S&P CNX NIFTY CNX NIFTY JUNIOR
Constant Coeff.
Std. Err.
t-stat.
Sig.
.203
.155
1.310
.191
.238
.195
1.220
.223
1 Monday Coeff.
Std. Err.
t-stat.
Sig.
beta
.076
.071
1.060
.290
.044
-.022
.079
-.281
.779
-.012
2 Tuesday Coeff.
Std. Err.
t-stat.
Sig.
beta
.043
.098
.438
.662
.018
.077
.098
.780
.436
.032
3 Wednesday Coeff.
Std. Err.
t-stat.
Sig.
beta
-.020
.093
-.220
.826
-.009
-.005
.096
-.055
.957
-.002
4 Thursday Coeff.
Std. Err.
t-stat.
Sig.
beta
.048
.095
.507
.612
.021
.235
.104
2.269
.024
.094
5 Friday Coeff.
Std. Err.
t-stat.
Sig.
beta
.060
.083
.715
.475
.030
.067
.091
.740
.460
.031
F-statistics .453 1.353
R Square 0.004 0.011
The size of the coefficient for each independent variable indicates the size of the effect that the
independent variable has on the dependent variable, and the sign on the coefficient (positive or negative)
gives the direction of the effect.
A t-stat of greater than 1.96 with significance less than 0.05 indicates that the independent variable is a
significant predictor of the dependent variable within and beyond the sample.
The greater the t-stat the greater the relative influence of the independent variable on the dependent
variable. To have a very large t-statistic implies that the coefficient was able to be estimated with a fair
amount of accuracy. If the t-stat is more than 2 you would generally conclude that the variable in
question has a significant impact on the dependent variable.
The above results show that all the value of t-stats are less than 2 for CNX NIFTY and CNX NIFTY JUNIOR
except on Thursday for CNX NIFTY JUNIOR. It indicates that the t-stat values coefficients are not having
significant impact on the dependant variable proving the non existence of weekend effect. Marquering,
Nisser, and Valla (2006) have shown that seasonal anomaliessuch as the weekend effectdiminish
soon after their discovery.
Results show that whenever there is a increase in the value of t-stat, simultaneously the value of the beta
has gone up showing a direct proportional relationship among each other for both the CNX NIFTY as
well as for the CNX NIFTY JUNIOR. Since t- statistic has no significant effect, it can be said that same
with the case of beta also.
The value of the R square signifies the proportion of variance in the dependent variable (i:e. return of the
weekdays) which can be explained by the independent variable (days of the week i:e Monday, Tuesday,
Wednesday, Thursday, Friday). This is an overall measure of the strength of association and does not
reflect the extent to which any particular independent variable is associated with the dependent variable.
The R square values are 0.004(.4 percent) and 0.011(1.1 percent) for the CNX NIFTY and CNX NIFTY
JUNIOR respectively which denotes that in CNX NIFTY.4% of the variation in the daily stock return is
explained by the closest previous days return and in case of CNX NIFTY JUNIOR only 1.1% of the
variation in the daily stock return is explained by the closest previous days return. Since both the values
are close to zero, the previous days stock price is a bad predictor of the current days stock price.
It is apparent from the study that since R square value is very low, this may be a sign that the stock
market has turned into a more efficient market over time. Vergin and McGinnis (1999), Keef and Roush
(2005), and Chong et al (2005) document a noticeable decline in weekend effect.

Conclusion:
The results and analysis of the study shows that there are lowest mean return on Mondays but
highest returns on Wednesdays. Hence, Monday effect is not recorded moreover to buy on
monads cannot be recommended since, Monday returns reports the maximum volatility. The
day-of-the-week effect may be present because surprisingly Wednesdays have yielded the
maximum returns across both the indices, but then the seasonality results indicate that there are
no significant days of the week effect in Indian stock market. So, the specific trading rule that
could be conceived of is that one could consider buying the scrips on Monday (buy low) and
selling them on Wednesday (sell high). However, this strategy needs to be exercised with
caution.