Pak J Commer Soc Sci

Pakistan Journal of Commerce and Social Sciences

2013, Vol. 7 (2), 381-393

Investigating Day-of-the-Week Effect in Stock

Returns: Evidence from Karachi Stock Exchange Pakistan
Muhammad Arshad Haroon
Assistant Professor , Government Sindh College of Commerce Hyderabad
Adjunct Assistant Professor, Isra University Hyderabad, Pakistan
E-mail: [email protected]
Nida Shah
Department of Management Sciences, Isra University, Hyderabad, Pakistan
E-mail: [email protected]
Abstract
This paper investigates Day-of-the-Week Effect in stock returns in the primary equity
market Karachi Stock Exchange (KSE) of Pakistan by employing OLS regression
approach. Data consists of daily closing prices of KSE-100 Index from January 01, 2004
to December 30, 2011. A traditional method of finding Day-of-the-Week Effect has been
comprised of only one regression equation. Contrary to this plausible methodology, this
paper proposes five separate models to statistically find significant effect on each trading
day of the week. Non-parametric Kolmogorov-Smirnov (K-S) test confirms abnormal
distribution of returns. Robust Standard Error addresses heteroscedasticity of returns;
proved by abnormal distribution. The t- statistics tests significance of coefficients
and One Factor ANOVA tests the hypotheses related to significant difference of
mean returns. Findings conclude mixed results due to the effect of political instability
on the anomaly. No effect found in Sub Period I. While, negative Monday and Positive
Friday effects revealed in Sub Period II; result consistent with the findings of Fields
(1931), Cross (1973), French (1980) and Haroon (2005).
Keywords: Karachi Stock Exchange; Day-of-the-week effect; KSE-100 index; OLS
regression.
1. Introduction
Stock markets are considered as one of the major component of financial system of the
capitalistic world. They provide place for the trading of financial assets such as
stocks and bonds of joint stock companies, guilt age securities, unit trusts and other
financial products efficiently, systematically, and by protecting the interest of investors.
Stock market functions as bridge and enables individuals as well as institutions to add in
countrys wealth through their participation in the secondary market of the country. A
well-regulated stock market promotes the phenomenon of fair pricing of securities and
reduces transactional costs. It stimulates economic growth, promotes economic activity
and an ample source of increasing employment in the country. A good performance of

Day-of-the-Week Effect in Stock Returns

stock market is a strong indicator of healthy economy. Because of aforesaid
reasons, stock traders keenly observe any single movement of stock index which may
affect their future profitability or help them to evaluate their portfolios. They also keenly
observe the economy; any sudden incident or change that may affect their decisions of
buying and selling stocks. Three stock exchanges are currently functioning in Pakistan
namely Karachi Stock Exchange, Islamabad Stock Exchange and Lahore Stock
Exchange.
Calendar anomaly or Seasonality is the most puzzling topic in finance on which
numerous studies have been conducted during the last three decades. Some of these
stock anomalies are Day-of-the-Week Effect, Weekly Effect, Month-of-the-Year Effect,
Weekend Effect, Ramadan Effect and January Effect etc. The most interesting of them
is Day-of-the-Week Effect which has attracted attention of researchers around the world.
Day-of-the-Week Effect refers to the observations that mean stock returns are
differently distributed among different week days. This phenomenon contradicts the
weak form of efficient market hypothesis given by Fama (1970) which says that; all
relevant information available to the market participants should not allow them to earn
abnormal returns (Marquering W. 2002). More specifically, weak form of Efficient
Market Hypothesis suggests that stock prices and stock returns should be normally
distributed. In contrast, the evidence collected from studies suggests the presence of
seasonal or calendar anomalies and thus stock returns are not constant.
2. An Overview of Karachi Stock Exchange (KSE)
Karachi Stock Exchange is known as the premiere and the most liquid stock exchange
of Pakistan with over 592 companies listed (divided into 33 different sectors). It is wellknown as the second oldest stock exchange in South Asia (Bashir et al 2011).World
famous magazine Business Week declared it the Best Performing Stock Market of
the World for the year 2002 (Business Week Magazine). In 2008, it was declared the
best performing emerging market (Gulf News 2008). KSE is regulated by SECP. Rules
and regulations are issued with a view to protect investors interest ensuring safe and
fair dealing with efficiency in the market. On June 30, 2011 the KSE-100 Index
produced a return of 28.5% from 9722 to 12496 on the basis of which KSE secured
fifth rank amongst 12 emerging Asian countries in the list of best performing
equity markets (KSE Annual Report 2011).
The objective of this study is to explore Day-of-the-Week Effect in Karachi Stock
Exchange by employing new methodology. A traditional method of finding Day-of-theWeek Effect has been comprised of only one regression equation. This potential
drawback served as motivation for this paper. The rest of this paper will follow the
following pattern: Section 3 elaborates scope of the study, Section 4 discusses literature
review, section 5 defines data, section 6 elucidates hypotheses, section 7 elaborates the
research methodology, section 8 reveals analysis and results, Section 9 discusses the
conclusion of the study, and section 10 enlists the literature cited for the study.
3. Scope of the Study
This study is based on the KSE-100 Index data from January 2004 to December 2011.
During the period under review there were various anomalies, economic and political
events, legislative changes and structural reforms that have affected the Pakistani equity
markets. In this study we compare two political periods: one Musharrafs regime
382

Haroon and Shah

from 2004 - 2007 and other democratic government 2008 - 2011. The period I shows
great political uncertainty. Prime Ministers of Pakistan kept on changing as well the
country was ruled by a Militant President for the first time in history. Though, it was
seen as a period of unprecedented economic growth. This era evidenced optimum level
reserves, growth in GDP, high foreign investment, no requirement of IMF funds.
Pakistani stock markets offered more buoyancy and dynamism. However, the last year
of period, 2007 constituted some bad events emergency rule imposed by Pervaiz
Musharraf and stock market faced biggest one day decline; KSE-100 Index fell nearly
5%.
Although, Period II comprised of a period of democratic and political stability but it
confronted increased violence, political fights, suicide bombing, extensive flood, raised
militancy along western borders of the country, skyrocketing inflation and gigantic
budget deficits, huge IMF Loans, energy crises, accelerating unemployment, and
climbing oil prices in international market. All these resulted in the steep decay of the
stock market. However the market rebounded strongly in 2009 and the trend continued in
2011. As we know that investment pattern vary extensively due to consistently changing
expectations of investors and participants about economically beneficial strategies.
Investors, in order to get more benefits, are interested in investing or withdrawing on any
particular day of the week. In this paper we are going to answer that question: Is
there any day of the week on which investor can get more benefit by investment
or withdrawal?
4. Literature Review
Day-of-the-Week Effect proposes the stock returns vary among different trading days
of the week. The First day of the week is usually considered as a week day because the
market remains bearish, while on the last day of the week the market is found bullish.
It can be supported with the explanation given by Mehdian and Perry (2001), that
unfavorable news mostly announced during the weekend shake the confidence of the
investors; stimulate them to sell their holdings on the coming Monday. Increase in
supply of the stocks cause prices to fall and resultantly return declines. Some
researchers elucidated the reason for this anomaly is the investors psychology; due to
the announcement of bad news on the weekend investors feel pessimistic on the first
trading day, whereas they feel optimistic on the last trading day and proceed on sales
and purchases. Nevertheless, the above explanations are not enough to adequately
explain the phenomenon which makes this anomaly the subject for continuous research.
This section outlines the relevant work done by different academicians, researchers and
investigators: Some of the pioneers who have contributed on this particular area include
Cross (1973) and French (1980), studied 50 shares of Standard & Poors
Composite Index to find Day-of-the-Week Effect and claimed higher mean returns on
Friday and lower mean returns on Monday. Whereas, Berument and Kiymaz (2001)
reported highest and lowest returns on Wednesday and Monday respectively by studying
shares of Standard & Poors Composite Index using OLS and GARCH Model. In
addition, volatility was highest on Friday and lowest on Wednesday.
Gibbons and Hess (1981) and Keim and Stambaugh (1984) studied the Dow Jones
Industrial Index and found negative Monday returns. Keim also described the positive
correlation between Friday and Monday returns. Similarly, Draper and Paudyal (2002)
383

Day-of-the-Week Effect in Stock Returns

conducted research on London stock exchange by using OLS and robust regression
procedure. They explained strong negative Monday returns, indicating the Monday
effect initiated by various factors. Day-of-the-Week Effect in Kuwait stock exchange
was investigated by Al- Mutairi (2010) using the three GARCH Models: GARCH, EGARCH and T-GARCH found higher returns on Saturday which is first day of the
week. Similar results were observed by Ulussever et al (2011) in TADAWAL stock
exchange.
Various other investigators have studied on different markets to observe Day of the
week anomaly. Some of these studies include: Jaffe and Westerfield (1985) found
weekend effect in stock markets of four developed countries: Australia, Japan, UK and
Canada. Aggarwal and Rivoli (1989) revealed the presence of strong Tuesday effect in
four Asian emerging markets: Hong Kong, Singapore, Malaysia and Philippines.
Kiymaz and Berument (2003) evaluated different international markets and found
returns are highly volatile on Monday for Germany and Japan, on Thursday for United
Kingdom and on Friday for Canada and United States. Similarly, Basher and Sadorsky
(2006) analyzed the closing returns of major stock indexes of 21 emerging stock
markets and the Morgan Stanley Capital International (MSCI) World index by
employing regression model adjusted for the market risk and volatility in returns. The
results showed negative Tuesday effect and positive Friday effect in Pakistan and
Taiwan respectively, Philippines has both Tuesday and Friday effect. No significant
results found in other countries. Furthermore, Apolinario et al. (2006) used GARCH
and T-ARCH models to analyze Day-of-the-Week Effect on major European stock
markets and concluded that most European markets do not reflect Day-of-the-Week
Effect except French and Swedish markets although volatility of returns in specific
markets followed similar behavior.
Several studies found no significant proof of Day-of-the-Week Effect in different stock
exchanges. Demirer and Karan (2002) studied Istanbul Stock Exchange by applying
regression equation on the unadjusted returns and on returns adjusted for inflation;
found no Day-of-the-Week Effect. They added that Start of the week is the good
indicator of behavior of market during the whole week. Agathee (2008) found no
significant results on SEMDEX. Similarly no Day-of-the-Week Effect was unveiled by
Lin and Yeh (2011) among eight major industries in Taiwan stock exchange.
A brief review of studies conducted to investigate Day-of-the-Week Effect in emerging
market of Pakistan is as follows: Khilji (1994) observed monthly returns are time
dependent. Haroon (2005) provided evidence of Monday or week day effect by
rejecting the weak form efficiency in KSE. In contradiction, Husain (1999) studied KSE100 Index and found no Day-of-the-Week Effect in Pakistani equity market. Later on,
Nishat and Mustafa (2002) reported no weekly effect on returns and conditional variance
in Pakistani stock markets by examining KSE-100 Index from 1991 to 2001. Mahmood
(2007) also evidenced the presence of random walk in Karachi Stock Exchange.
Furthermore, Ahmed and Roser (1995) clarified that Pakistani equity market displays
complex dynamics due to presence of speculative bubbles. Ali and Akbar (2009) studied
Pakistani stock market by using AR1 and OLS regression model on returns data of
KSE-100 share Index from November 1991 to October 2006. They found no weekly
effects or monthly effects in stock returns.
Some recent studies include: Husain et al (2011) used OLS regression and concluded
384

Haroon and Shah

that stock market returns for Tuesday are higher and more volatile than other days
of the week. Bashir et al (2011) refuted Efficient Market Hypothesis in banking
sector of Pakistan. They studied the daily closing stock prices of 11 major trading
banks listed on KSE by using the Co-integration and VAR test.
The point of attention in the above literature is the employment of only one regression
equation to find several Day-of-the-Week Effects. This obvious shortcoming influenced
to enrich the literature with new methodology by giving possible explanation and insight
about this phenomenon, primarily focusing Pakistani equity market.
5. Data
The data comprises of daily closing price of KSE-100 index from January 01, 2004 to
December 30, 2011 except official holidays. It is pointed out here that KSE operates
five days in a week from Monday to Friday. We calculate natural log of daily returns in
order to reduce the effect of size of daily stock prices. The daily return is calculated
by using the following formula:

Sub Periods
Sub Period I
Sub Period II

Table 1: Division of Timeline

Duration
No. Of observations
Jan 01, 2004 Dec 31, 2007
967
Jan 01, 2008 Dec 30, 2011
890

Rx = (ln KSE 100 x - ln KSE 100 x-1) * 100

Where, Rx is the return on KSE-100 Index on day x, KSE 100x is the closing price
of KSE-100 Index on day x and KSE 100x-1 is the closing price of KSE-100 index on
day x-1. Data has been collected from database of Karachi Stock Exchange. The whole
sample period consists of 1857 observations. For the purpose of study we divide our data
in two equal Sub Periods. The necessary detail is given in Table 1.
6. Hypotheses
Following hypotheses has been developed for testing Day-of-the-Week Effect on
returns in KSE:
H1: The mean stock returns across all trading days of the week are significantly
different and Day-of-the-Week Effect is present in Sub Period I (2004 2007).
H2: The mean stock returns across all trading days of the week are significantly
different and Day-of-the-Week Effect is present in Sub Period II (2008 2011).
7. Research Methodology
Ordinary Least Squares (OLS) Regression is the common approach employed by the
previous studies to investigate Day-of-the-Week Effect in stock returns. But their
shortcoming lies in utilizing only one regression equation to find effect on all trading
days of the week. This approach is possible only if we hold a prior belief that an effect
exists on one specific day, such as Monday. However, this specification is not appropriate
if we have no previous expectation on which of the Day-of-the-Week Effect might exist.
We overcome this shortcoming by estimating a different model for finding effect on each
trading Day of the Week by omitting the dummy variable for the day under scrutiny in
each case Borges (2009).
385

Day-of-the-Week Effect in Stock Returns

Monday Effect:

RMx = + 2Tx + 3Wx + 4THx + 5Fx + ex

Tuesday Effect:

RTx= + 1Mx + 3Wx + 4THx + 5Fx + ex

Wednesday Effect:

RWx = + 1Mx + 2Tx + 4THx + 5Fx + ex

Thursday Effect:

RTHx = + 1Mx + 2Tx + 3Wx + 5Fx + ex

Friday Effect:

RFx = + 1Mx + 2Tx + 3Wx + 4THx + ex

Where,

RMx , RTx, RWx, RTHx and

RFx =
=

--

Index returns on day Monday, Tuesday,

Wednesday, Thursday and Friday respectively.
Mean return on respective day
OLS coefficients which represents the
difference between the mean return for the
particular day (Monday, Tuesday, Wednesday,
Thursday and Friday) and the mean return for
that day under scrutiny in each case respectively.

DMx, DTx, DWx, DTHx,

DFx =

Dummy variables for Monday, Tuesday,

Wednesday, Thursday and Friday respectively;
DMx, = 1 if x is Monday; otherwise 0 for all
other days, DTx = 1 if x is Tuesday; otherwise
0 for all other days and so forth.

ex

Robust Standard Error.

Daily return data is distributed abnormally which proves heteroscedasticity in

returns; thus requires Robust Standard Error in linear regression model (Borges
2009). Robust Standard Error is the White's heteroscedasticity consistent standard
errors. To prove Day-of-the-Week Effect on any day (1) the mean return of the day
should be significant and (2) at least two of the excess returns should be significantly
different from the mean return of the day under scrutiny.
8. Analysis and Results
8.1 Descriptive Statistics
Descriptive statistics of the KSE-100 Index returns for whole sample period is reported
in Table 2.

386

Haroon and Shah

Table 2: Descriptive Statistics for Daily Return For Sample Period

Sample Period
January 01, 2004 to December 30, 2011
Observations (N)
1857
Maximum Return
10.68888
Minimum Return
-0.42428
Mean Return
0.050149
Standard Deviation
1.812417
Variance
3.285
Skewness
-7.205
Std. Error
0.057
Kurtosis
164.983
Std. Error
0.114
Table 2 summarizes the basic Descriptive Statistics of KSE-100 Index returns for entire
sample period. The skewness and kurtosis of total returns demand attention. Skewness
suggests data is tailed towards smaller values and it is left or negatively skewed. High
value of kurtosis indicates distribution is more clustered and thick tailed or leptokurtic.
(Demirer and Karan, 2002) reported that stock exchange returns often have leptokurtic
distribution. Moreover, returns are abnormally volatile on Thursday. The implications are
vivid by histogram of the returns and the K-S normality test; both rejecting the normality
of returns.
Figure 1 portrays the histogram of returns with normal distribution curve. The curve
looks much normal which suggests that the errors may be fairly normal. Since the value
of skewness and kurtosis are much more than their respective standard error values this
suggests a degree of abnormality in returns.

Figure 1: Histogram of Daily Returns (With Normal Curve)

387

Day-of-the-Week Effect in Stock Returns

Table 3 describes the results of non - parametric Kolmogorov-Smirnov (K-S)
G oodness of fit test. Probability of Z score at 5% significance level confirms the
abnormality of distribution.
Table 3: One Sample Kolmogorov-Smirnov Goodness of Fit Test
Sample
N
Normal Parameters a

1857
Mean

0.0501489

Standard Deviation
Absolute
Positive
Negative

Most Extreme Differences

1.81241711
0.136
0.107
-0.136
5.874
0.000

Kolmogorov-Smirnov Z
Asymp. Sig. (2-tailed)

a. Test distribution is Normal.

The basic statistics of the returns of two Sub Periods for each day of the week are
presented in Table 4.
Table 4: Summary Statistics For Daily Return For Sub Periods
Sub Period I
Mon
Tues
Wed
Thur
Fri
N
Max. Return
Min. Return
Mean Return
P Value
Std. Deviation
Variance
Skewness
Std. Error
Kurtosis
Std. Error

193
10.6889
-4.8280
-0.062528
0.671
2.0431
4.174
0.789
0.175
4.820
0.348
Mon

N
Max. Return
Min. Return
Mean Return
P Value
Std. Deviation
Variance
Skewness
Std. Error
Kurtosis
Std. Error

188

193

3.9573
4.2498
-4.5342
-6.0624
0.250465
0.282602
0.021
0.005
1.4703
1.3934
2.162
1.942
-0.794
-0.692
0.177
0.175
1.709
2.717
0.353
0.348
Sub Period II
Tues
Wed

174

180

180

5.4056
-4.5346
-0.270874
0.018
1.5006
2.252
-0.224
0.184
1.891
0.366

8.2547
-12.8890
0.076235
0.543
1.6767
2.811
-1.821
0.181
22.031
0.360

388

4.7725
-5.1349
0.129554
0.230
1.4421
2.079
-0.373
0.181
2.250
0.360

197

195

4.5758
-4.4799
0.076064
0.465
1.4587
2.128
-0.570
0.173
2.262
0.345

4.2157
-4.2902
0.102683
0.267
1.2888
1.661
-0.625
0.174
1.968
0.346

Thur

Fri

181

175

4.7032
-42.4288
-0.244521
0.331
3.3780
11.411
-10.908
0.181
136.742
0.359

4.3185
-4.7243
0.187450
0.050
1.3168
1.734
-0.197
0.184
2.991
0.365

Haroon and Shah

In Sub Period I the mean return for Monday is lowest and Wednesday has the highest
mean return this result is similar to one found by Berument and Kiymaz (2001) studied
S&P 500 stock index. Monday shows lowest mean return and Friday shows highest
mean return in Sub Period II. These results are consistent with Fields (1931), Cross
(1973) and French (1980).
Another substantial phenomenon is Volatility. In Sub Period I, Monday returns are most
volatile as it possesses highest variance, skewness and Kurtosis. In Sub Period II,
Thursday returns show highest variance, skewness and Kurtosis. Another difference
between both Sub Periods is that Sub Period II has high variance, skewness and kurtosis
values than Sub Period I. High skewness suggests a less developed market moving
asymmetrically to 'news' and huge kurtosis provides that movements of share prices are
very large. Gupta (1997) termed these movements the result of "euphoria to despondency
cycles".
8.2 OLS Regression Results
Table 5 and table 6 exhibit the estimates of parameters along with t- statistics for Sub
Period I and II respectively for each day effect. The significance is considered at 5%
level. F Value is also calculated by testing model for each day effect through ANOVA
at 5% significance level.
OLS results for Sub Period I negate the presence of Day-of-the-Week Effect since most
of the coefficients are not significant from each other, except one in Thursday and
Friday effect; which is insufficient for drawing any conclusion. Furthermore, ANOVA
suggests all the five models are insignificant because the F significance value for all
models is higher than 0.05. Therefore the null hypothesis is accepted with the conclusion
that the mean stock returns across all trading days of the week are not significantly
different and Day-of-the-Week Effect not is present in Sub Period I (2004 2007).
These Findings are consistent with Ahmed and Roser (1995), Husain (1999), Nishat and
Mustafa (2002), Mahmood (2007), Ali and Akbar (2009). The results are shown in table
5.

389

Day-of-the-Week Effect in Stock Returns

Table 5: OLS Regression Results of Sub Period I

Sub Period I
(2004 2007)
Monday Effect

-0.036

-0.101

0.094

-0.038

-0.054

0.527

t - stat.

-0.99123

0.883836

-0.362

-0.46243

Test Insignificant

3
0.033
0.4297

4
0.079
1.0605

5
-0.029
-0.347

F Value
P Value
0.641
0.634
Test Insignificant

2
0.031
0.430

4
-0.031
-0.434

F Value
P Value
0.382
0.822
Test Insignificant

2
0.077
1.060

3
-0.033
-0.43399

5
-0.188
-2.305*

F Value
P Value
1.721
0.147
Test Insignificant

2
-0.023
-0.34723

3
-0.045
-0.67138

4
-0.150
-2.30453*

F Value
P Value
1.545
0.191
Test Insignificant

Tuesday Effect

1
0.0234
-0.053
-0.9912
t - stat.
Wednesday Effect

1
0.2814
0.045
0.884
t - stat.
Thursday Effect

1
0.099
-0.19
-0.362
t - stat.
Friday Effect

1
0.121
-0.022
-0.46243
t - stat.

F-Value

-0.054
-0.671

P-Value
0.716

In Sub Period II, three significant coefficients found in Monday and Tuesday effect;
whereas, two in Wednesday, Thursday and Friday effect. Moreover, ANOVA
recommends the F-values for all the five models found significant. This depicts the
presence of Day-of-the-Week Effect. Therefore the null hypothesis has been rejected and
it is concluded that the mean stock returns across all trading days of the week are
significantly different and Day-of-the-Week Effect is present in Sub Period II (20082011). Further, there is conclusive evidence of significant negative Monday returns and
positive Friday returns; since it shows lowest and highest significant mean returns on
Monday and Friday respectively. It leads to the conclusion that Negative Monday and
Positive Friday effect is present in KSE in Sub Period II (2008 2011). These results
are consistent with the findings of Fields (1931), Cross (1973), French (1980) and with
Haroon (2005). Whereas, sharply contrasts the study on Day-of-the-Week Effect in KSE
by Ahmed and Roser (1995), Husain (2000), Nishat and Mustafa (2002), Mahmood
(2007), by Basher and Sadorsky (2006), Ali and Akbar (2009), and Husain et al
(2011). The results are vivid from table 6.

390

Haroon and Shah

Table 6: OLS Regression Results of Sub Period II

Sub Period II
(2008 2011)
Monday Effect

-0.036

-0.101

0.094

-0.038

-0.054

t - stat.

-0.99123

0.883836

-0.362

-0.46243

Test Insignificant

3
0.033
0.4297

4
0.079
1.0605

5
-0.029
-0.347

F Value
P Value
0.641
0.634
Test Insignificant

2
0.031
0.430

4
-0.031
-0.434

5
-0.054
-0.671

F Value
P Value
0.382
0.822
Test Insignificant

2
0.077
1.060

3
-0.033
-0.43399

5
-0.188
-2.305*

F Value
P Value
1.721
0.147
Test Insignificant

2
-0.023
-0.34723

3
-0.045
-0.67138

4
-0.150
-2.30453*

F Value
P Value
1.545
0.191
Test Insignificant

Tuesday Effect

1
0.0234
-0.053
-0.9912
t - stat.
Wednesday Effect

1
0.2814
0.045
0.884
t - stat.
Thursday Effect

1
0.099
-0.19
-0.362
t - stat.
Friday Effect

1
0.121
-0.022
-0.46243
t - stat.

F Value

P Value

0.527

0.716

9. Conclusion
This study proposed new methodology to find Day-of-the-Week Effect. Five OLS
regression equations were built to find each Day-of-the-Week Effect as suggested by
(Borges, 2009). We mainly focus on KSE-100 Index of Karachi Stock Exchange but the
methodology possesses applicability in any stock market. The study found mixed
results for both Sub Periods. Sub Period I (2004 2007) negates the prevalence of
Day-of-the-Week Effect due to political instability. Moreover, study evidenced the
presence of Day-of-the-Week Effect in Sub Period II. The argument to support this
result might be that the period after election in 2008 - 2011 comprised of democracy
and political stability in terms of government and fairly consistent policies.
OLS regression for Period II revealed a negative Monday and Positive Friday effect.
These results are consistent with findings of Fields (1931), Cross (1973), French (1980).
Finally, we conclude that investors can get more benefit by investment or withdrawal on
Monday and Friday.
In future, research may be conducted on Individual securities as well as industry
groups. Moreover, seasonality may also be tested by considering the volatility pattern in
returns.
Models for heteroscedasticity like ARCH, GARCH, T-GARCH and EGARCH may be employed to find more significant results. Further researches must be
conducted to find the impact of politics and other fundamental factors on anomalies
in stock returns. I s u g g e s t t h a t t he institutions and brokers who provide financial
services should include seasonal anomalies in their analysis.
391

Day-of-the-Week Effect in Stock Returns

REFRENCES
Agathee, U.S. (2008). Day-of-the-Week Effects: Evidence from the Stock Exchange of
Mauritius (SEM). International Research Journal of Finance and Economics, 17, 7-14.
Aggarwal, R. and Rivoli, P. (1989). Seasonal and Day-of-the-Week Effects in Four
Emerging Stock Markets. The Financial Review, 24(4), 541-550.
Ahmed, E. and Roser, J.B. ( 1995). Nonlinear Speculative Bubbles in the Pakistani
Stock Market. Journal of Pakistan Development, 34(1), 25-41.
Ali, S. and Akbar, M. (2009). Calendar Effects in Pakistani Stock Market.
International Review of Business Research Papers, 5(1), 389-404.
Al-Mutairi, A. (2010). An Investigation of the Day-of-the Week Effect in the Kuwait
Stock Exchange. Research Journal of International Studies, 16, 191-196.
Apolinario, R.M.C., Santana, O.M., Sales, L.J., and Caro, A.R. (2006). Day-of-the-Week
Effect on European Stock Markets. International Research Journal of Finance and
Economics, 17, 53-62.
Basher, S.A. and Sadorsky, P. (2006). Day-of-The-Week Effects in Emerging Stock
Markets. Applied Economics Letters, 13, 621628.
Bashir, T., Ilyas, M. and Furrukh, A. (2011). Testing Weak Form-Efficiency of
Pakistani Stock Markets An Empirical Study in Banking Sector. European Journal of
Economics, Finance and Administrative Sciences, 31, 160175.
Berument, H. and Kiymaz, H. ( 2001). The Day-of-the-Week Effect on Stock Market
Volatility. Journal of Economics and Finance, 25(2), 181-193.
Borges, M.R. ( 2009). Calendar Effects in Stock Markets: Critique of Previous
Methodologies and Recent Evidences in European Countries. Working Paper 0874
4548 No.37/2009/DE/UECE, School of Economics and Management, Department
of Economics, Technical University of Lisbon.
Cross, F. (1973). The Behavior of Stock Prices on Fridays and Mondays. Financial
Analysis Journal, 29(6), 583-588.
Demirer, R. and Karan, M.B. (2002). An Investigation of the Day-of the-The-Week
Effect on Stock Returns in Turkey. Emerging Markets Finance & Trade, 38(6), 47-77.
Draper, P. and Paudyal, K. (2002). Explaining Monday Returns. Journal of Financial
Research, 25(4), 507520.
Fama, E . ( 1970). Efficient Capital Markets:
Work. Journal of Finance, 25(2), 383-417.

A Review of Theory a nd Empirical

Fields, M.J. (1931). Stock Prices: A Problem in Verification. The Journal of

Business of the University of Chicago, 4(4), 415-418.
Foley, T. (2008). Pakistan Emerges a Market Winner. Gulf News Investment Gulf
News, [Online] Available: http://gulfnews.com/business/investment/pakistan-emerges-amarket-winner-1.97437, (accessed on 13 March 2012)
French, K.R. (1980). Stock Returns and the Weekend Effect. Journal of Financial
Economics, 8(1), 55-69.
Gibbons, M.R. and Hess, P. (1981). Day-of-the-Week Effects and Asset Returns. The
392

Haroon and Shah

Journal of Business, 54(4), 579-596.

Gupta, L.C. (1997). India's Stock Market Crisis, Society for Capital Market Research and
Development. Research Monograph No. 5, Vivek Financial Focus Limited.
Haroon, M.A. (2005). Trend Analysis of Karachi Stock Exchange: A Case Study of
Ten Years (1992-2002). Doctoral Dissertation in Business Administration, Hamdard
University, Karachi.
Husain, F. (1999). The Day-of-the-Week Effect in the Pakistani equity market: An
Investigation. The Lahore Journal of Economics, 5(1), 93-98.
Husain, F., Hamid, K., Akash, R.S.I. and Khan, M.I. (2011). Day-of-the-Week Effect
and Stock Returns: Evidence from Karachi Stock Exchange-Pakistan. Far East Journal of
Psychology and Business, 3(1), 25-30.
Jaffe, J. and Westerfield, R. ( 1985). The Weekend Effect in Common Stock Returns:
The International Evidence. The Journal of Finance, 40(2), 433-454.
Keim, D. B., and Stambaugh R.F. (1984). A Further Investigation of the Weekend
Effect in the Stock Returns. The Journal of Finance, 39(3), 819-835.
Khilji, N.M. (1994). Non-linear Dynamics and Chaos: Application to Financial Markets
in Pakistan. The Pakistan Development Review, 33(4), 1417-1429.
Kiymaz, H. and Berument, H. (2003). The Day-of-the-Week Effect on Stock Market
Volatility and Volume: International Evidence. Review of Financial Economics, 12,
363-380.
KSE Annual Report (2011). Karachi Stock Exchange. [Online] Available: http://
www.kse.com.pk . (Accessed on 2 March 2012).
Lin, C. and Yeh, C. (2011). The day-of-the-week effect among industries- evidence
From Taiwan. Review of Global Management and Service Science, 1, 45-50.
Mahmood, H. (2007). Market Efficiency: An Empirical Analysis of KSE 100 Index.
Journal of Business Forthcoming, 5(4), 86-98.
Mangi, N.A. (2012). And Worlds Top Market is Pakistan. Business Week, 22 January
2012, [Online] Available:
h t t p: / / www.businessweek.com/bwdaily/dnflash/apr2003/nf20030410_4232_db039.htm
Marquering, W. (2002). Seasonal Predictability of Stock Market Returns. Review of
Business and Economics, XLVII(4), 557-576.
Mehdian, S. and Perry, M.J. (2001). The Reversal of the Monday Effect: New
Evidence Form US Equity Markets. Journal of Business Finance and Accounting, 28,
1043-1065.
Nishat, M. and Mustafa, K. (2002). Anomalies in Karachi Stock Market: Day-of-theWeek Effect. The Bangladesh Development Studies, XXVIII(3), 55-64.
Poshakwale, S. (1996). Evidence on weak form efficiency and Day-of-the-Week Effect
in Indian Stock Market. Finance India, 10(3), 605-616.
Ulussever, T., Yumusak, I.G., Kar, M. (2011). The Day-of-the- Week Effect in the
Saudi Stock Exchange: A Non-Linear Garch Analysis. Journal of Economic and Social
Studies, 1(1), 9-23.
393