INTRODUCTION
Market Efficiency influences the Investment Strategy for investors because
in an efficient market, there would be no undervalued or overvalued stocks.
The efficient market theory states that an informationally efficient market
is one where the market price is an unbiased estimate of the true value of the
investment. It further states that the current market price of a security fully
reflects all available information and the current price is the fair price as
the security has traded in that price (Fama et al.,
1969). In the words of Fama, “the informational efficiency of financial
markets requires that the market prices and rates of return at any given time
reflect all the information available to the participants” (Fama,
1970).
The academics and practitioners have documented many research works on the
Seasonality and associated behavior of securities markets all over the world.
Among others, the most widely mentioned Seasonal Effects and market anomalies
are January effect, Monday effect or weekend effect, holiday effect and small
firm effect, to mention a few. Among these, one of the widely discussed anomalies
is the Monday stock return. The most common case is the Monday effect, meaning
that the Monday’s average return is significantly lower than the other
days’ average returns. Fridays normally present the highest return over
majority of the stock markets of the world. However, some empirical studies
in different stock markets have established the Tuesday Effect instead of the
Monday Effect. During the past decades, many studies about the Day of the Week
Effect have been carried out.
It is significant to note that there is a reason for the day of the week effect.
Monday recorded high return in some markets. In another, Monday recorded lower
return. The reason is that Monday is the day with the lowest trading volume
and in which the propensity of individuals to transact is higher relative to
other days of the week and that of the institutions is the lowest. The propensity
of individuals to sell on Monday is higher than their propensity to buy (Lakonishok
and Maberly, 1990).
The other reason is that the Settlement Cost has been used to explain day of the week variations. There are five trading days in a stock market. If the settlement day is the second trading day, the Thursday return will be higher than rest of the week days. If the investors buy on the Wednesday’s close price and sell on the Thursday’s close price, then investors will earn high return on Thursday. Another main reason is the individual investors’ behavior. The individual investors would like to sell more on Monday due to the reason that the bad news is normally released in the prior week and the individual investor tends to use Monday as the opportunity to satisfy the liquidity needs. It is hard to say that the day of the week effect can generate abnormal returns. It is always possible to find the abnormal returns for short periods but it seems a much harder task to generate abnormal returns over a longer period, as Anomalies vary over time and tend to disappear or even reverse after they have been discovered.
Ravi and Goswami (2000) studied the weekend effects
by using equally weighted portfolio constructed from 70 stocks listed on the
BSE. The study evidenced the (heteroskedasticity adjusted) excess positive returns
on Friday and excess negative returns on Tuesday. Amanulla
and Thiripalraiu (2001) proposed to find out whether the carryforward transactions
in different periods have any impact on weekend effect in the Indian Stock
Market. The results from the subsample period strongly supported the existence
of weekend effect during the period of ban on carry forward (badla) transactions.
This study also evidenced a reversal in WeekEnd Effects, i.e., positive Monday
return and negative Friday return in modified and revised modified carry forward
transactions. Brooks and Persand (2001) examined the
evidence for the Day of the Week Effect in five Southeast Asian Stock MarketsTaiwan,
South Korea, the Philippines, Malaysia and Thailand. The Authors found that
neither South Korea nor Philippines recorded significant Calendar Effects. But
both Thailand and Malaysia registered significant positive average returns on
Monday and significant negative average returns on Tuesday. In addition, the
study also documented a significant negative Wednesday Effect in Taiwan. Nath
and Dalvi (2005) used both high frequency and end of day data for the benchmark
index (S and P CNX Nifty). The study, using Regression with biweights and dummy
variables, found that before the introduction of Rolling Settlement in January
2002, Monday and Friday were significant days. However, after the introduction
of the Rolling Settlement, Friday has become significant. The market inefficiency
still exists and the market was yet to price the risk appropriately. Basher
and Sadorsky (2006) used both unconditional and conditional risk analysis
to investigate the dayoftheweek effect in 21 emerging stock Markets. The
results of this study showed that while the DayoftheWeek Effect was not present
in the majority of Emerging Stock Markets studied, some Emerging Stock Markets
did exhibit strong dayoftheweek effect even after accounting for conditional
market risk. Kumar and deo (2007) analyzed the efficiency
of Indian stock market by using S and P CNX 500 Index. The study found the presence
of Day of the Week Effect in the Indian Stock Market which affected both the
stock returns and volatility, thereby proving the Indian Stock Market to be
inefficient. Elango and AlMacki (2008) investigated
whether the anomalous Week End Effect was found in the rapidly emerging Indian
Equity Market. Their analysis produced mixed results, indicating that the Monday
Returns were negative and low in the case of two out of three indices. The study
also examined the Week End Effects and showed that Monday Returns were negative
in one of the bench mark indices. Nageswari and Babu (2011)
examined the Week End Effect in the Indian Stock Market. The study found that
the mean returns were positive for all days of the week, highest on Friday and
lowest on Monday. It was inferred that the Day of the Week Pattern did not exist
in the Indian Stock Market during the study period. Nageswari
and Selvam (2011) explored the Day of the Week Effect during the Post Rolling
Settlement Period. The study found that the Highest Mean Return on Friday and
the Lowest Mean Return on Tuesday were observed during the study period. Further,
there was strong significant positive relationship between MondayFriday and
no significant relationship among other days of the week. The results indicated
that the Day of the Week Effect did not exist in the Indian Stock Market during
the study period.
The above literature provides an overview of Day of the week Effects in various Global Stock Markets. It is to be noted that only few have focused on the Monday Effect in the Indian Stock Markets. Against this backdrop, this study makes an attempt to examine whether India which is one of the fast emerging markets, offers evidences of Anomaly, thus ensuring abnormal returns to the investors.
Firms and Governments generally release good news between Monday and Friday and bad news on the weekends. As a result, the bad news is reflected in lower stock prices on the next trading day (Mondays) and good news is reflected in higher stock prices on Friday. This would reduce the share price further. Similarly, in the Month of January, firms normally release new information pertaining to the previous accounting year. When new positive information reaches the market, the prices become bullish due to buying pressure. The active trading strategies, based on the knowledge of market anomalies, would provide benefits to the investors. But the countervailing arbitrage will also exploit the excess return over time. In this environment, it is necessary to periodically find out whether these types of Anomalies exist in the Stock Market. Against this background, the present study on Monday Effects in the Indian Stock Market is significant.
The present study intends to identify and analyze the Monday Effect in the Indian Stock Market.
The present study tested the following null hypothesis: NH1: There are no significant differences among the returns of different trading days of the week.
MATERIALS AND METHODS
Sample selection: The indices are the best indicator of the performance
of the whole economy. The S and P CNX Nifty is well diversified, with 50 stocks
accounting for 22 Sectors of the Economy. It represents about 56% of the Free
Float Market Capitalization as on September 30th, 2010. The S and P CNX 500
is India’s first broad based benchmark of the Indian Capital Market. It
represents about 90% of the Free Float Market Capitalization and about 87% of
the total turnover on the NSE. For the purpose of this study, S and P CNX Nifty
and S and P CNX 500 Index were considered as Sample Indices.
Sources of data: The required information for the present study were
collected from the www.nseindia.com
and prowess which is a corporate database maintained by CMIE.
Period of the study: The present study covers a period of eight years from 1st April 2002 to 31st March 2010.
Tools used for analysis: The following tools were used for the analysis of the returns and volatility of the sample indices taken for this study.
Returns: The formula below was used to compute the daily returns for
each of the index series:
Where:
R_{t} 
= 
Daily return on the Index (I) 
ln 
= 
Natural log of underlying market series (I) 
I_{t} 
= 
Closing value of a given index (I) on a specific trading day (t) 
I_{t1} 
= 
Closing value of the given index (I) on preceding trading day (t1) 
Descriptive statistics: Under Descriptive Statistics, the Average Daily Returns (mean), Standard Deviation, Skewness and Kurtosis were used.
KruskallWallis test: The KruskallWallis Test is employed for testing
the equality of mean returns among different months of the year. The formula
for calculating the Test Statistic ‘H’ is as under:
Where:
Rj 
= 
Sum of the Ranks in the jth Column 
nj 
= 
Number of Cases in the jth Column 
N 
= 
Sum of Observations in all the Columns 
Dummy variable regression model: In order to investigate the Monday
Effect, the following dummy variable regression equation is used:
Where:
R_{i}t 
= 
The return of the Index on day t 
D_{1(Mon)} 
= 
Dummy variable equal to 1 if t is a Monday and 0 otherwise 
D_{2(Tue)} 
= 
Dummy variable equal to 1 if t is a Tuesday and 0 otherwise 
D_{3(wed)} 
= 
Dummy variable equal to 1 if t is a Wednesday and 0 otherwise 
D_{4(Thu)} 
= 
Dummy variable equal to 1 if t is a Thursday and 0 otherwise 
D_{5(Fri)} 
= 
Dummy variable equal to 1 if t is a Friday and 0 otherwise 

= 
Error term 
The intercept, β_{1}……..β_{5}, represent the average deviation of each day from the 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 (OLS).
RESULTS AND DISCUSSION
Analysis of descriptive statistics: Table 1 exhibits
the results of Descriptive Statistics for S and P CNX Nifty and S and P CNX
500 Index returns for the period from 1st April 2002 to 31st March 2010. The
above Table clearly observes that the S and P CNX Nifty Index returns recorded
the Highest Mean Return (0.1521) on Friday and the Lowest Mean Return (0.0175)
earned on Monday. This Anomaly be due to some unanticipated events or corporate
announcements that would have been reflected in the stock prices. The Standard
Deviation for the mean returns of S and P CNX Nifty Index ranged from 1.56 to
2.12% during the study period. The Highest Value (2.1226) of Standard Deviation
was recorded on Monday, with the least mean return and the Lowest Value (1.5687)
of Standard Deviation, earned on Tuesday. This indicates the fact that there
was nonlinearity between risk and return of S and P CNX Nifty Index in the
National Stock Exchange. In short, the market (NSE) was more volatile on Monday
and least volatile on Tuesday. The return distribution of S and P CNX Nifty
was positively skewed on Monday and Tuesday and negatively skewed on other days
of the week. During the study period, the result of kurtosis measure of Return
Distribution was Leptokurtic for all days of the week and highest (15.59) on
Monday. The reason for nonnormality of S and P CNX Nifty Index could be the
high kurtosis.
The analysis of S and P CNX 500 Index found that there were positive mean returns for all days of the week. The highest mean return (0.1701) was recorded on Friday and the lowest mean return on Monday during the study period. This study also found that there was the least Standard Deviation of the return recorded on Thursday and highest on Monday. It implies that the stock market (NSE) was more volatile on Monday and least volatile in Thursday during the study period. According to the analysis of S and P CNX 500 Index, the return distribution was positively skewed on Tuesday and negatively skewed for remaining days of the week. Regarding kurtosis measure, the Return Distribution of S and P CNX 500 was Leptokurtic for all days of the week and the highest value (14.21) was recorded on Monday during the study period.
Table 1: 
The results of descriptive statistics for S and P CNX Nifty
and S and P CNX 500 index daily returns from April 2002 to March 2010 

Source: Computed from PROWESS 
Analysis of KruskallWallis test: The results of KruskallWallis Test for S and P CNX Nifty and S and P CNX 500 Index Returns from 1st April 2002 to 31st March 2010 are presented in Table 2. As stated earlier, the KruskallWallis Test is commonly used to test the equality of mean returns of the different days of the week. The above Table clearly shows that the Value of KruskallWallis Test Statistic (S and P CNX Nifty2.29 and S and P CNX 500 Index5.28) was lower than the Table value (9.488) at 5% level of significance in 4 degrees of freedom for the sample Index returns. It clearly indicates that there was no significant difference between the returns of different days of the week.
Analysis of dummy variable regression model: Table 3 shows the results of the linear regression analysis for S and P CNX Nifty and S and P CNX 500 Index from April 2002 to March 2010. It is to be noted that the Benchmark Month in the Model was Monday, represented by the Intercept. The Values of Coefficients in Thursday was high and none of the variables was statistically significant at conventional level of risk in S and P CNX Nifty Index Returns. The above Table also reveals that the adjusted Rsquared value of 0.0058 was low. However, from the insignificant Fvalue, the Null Hypothesis, namely “There is no significant difference among the different days of the week” is not rejected. It indicates that the study did not confirm any Anomalies in S and P CNX Nifty Index during the study period.
The S and P CNX 500 Index returns recorded Positive Coefficient Value for all days of the week. It is to be noted that the value of coefficients (0.3983) on Thursday was high and statistically significant at 5% level. Hence the Null Hypothesis viz., “There is no significant difference among the different days of the week” cannot be rejected because the Fvalue was not statistically significant at conventional level of significance. In other words, there was no day effect in case of S and P CNX 500 Index returns during the study period. The adjusted Rsquared value of 0.0086 clearly indicates the fact that only 8.6% influenced these variables. Besides, Fstatistic indicates that the overall fit of the model was poor. Further, DurbanWatson Statistic of 1.78 indicates autocorrelation in the residuals.
Table 2: 
The results of KruskallWallis test for S and P CNX Nifty
and S and P CNX 500 index daily returns from April 2002 to March 2010 

Source: Computed from PROWESS using SPSS, Degrees of freedom.
N1 4, N = 5, Table value: 1%13.277, 5%: 9.488 
Table 3: 
The results of dummy variable regression model for S and P
CNX nifty and S and P CNX 500 index daily returns from April 2002 to March
2010 

Source: Computed from PROWESS using Eviews. *Significant
at 5% level 
Outcomes of the study: The following are the important findings of the present study:
• 
The analysis of the study reveals that there were Positive
Mean Returns recorded for all days of the week while the Highest Mean Returns
(0.1521 for S and P CNX Nifty, 0.1701 for S and P CNX 500) were recorded
on Friday and the Lowest Mean Returns (0.0175 for S and P CNX Nifty, 0.0306
for S and P CNX 500) were recorded on Monday for the sample indices during
the study period 
• 
It is suggested that the investors may buy the shares on Monday
and sell them on Friday because they may get returns better than on other
days of the week 
• 
The Standard Deviation for the mean returns of S and P CNX
Nifty Index ranged from 1.56 to 2.12% during the study period. The Highest
Value (2.1226) of Standard Deviation was recorded on Monday, with the least
mean return and the Lowest Value (1.5687) of Standard Deviation, earned
in Tuesday 
• 
This indicates the fact that there was nonlinearity between
risk and return of S and P CNX Nifty Index in the National Stock Exchange.
Hence, the Regulators may take necessary steps to maintain risk and return
tradeoff 
• 
According to S and P CNX 500 Index Returns, that there was
least Standard Deviation of the return recorded on Thursday and highest
on Monday. It implies that the stock market was more volatile on Monday
and least volatile in Thursday during the study period 
• 
The Return Distribution was Positively Skewed for Monday (0.114
for S and P CNX Nifty) and Tuesday (0.1231 for S and P CNX Nifty, 0.0042
for S and P CNX 500) while Negatively Skewed for remaining trading days
of the week during the study period 
• 
The Kurtosis measure of Returns Distribution was Leptokurtic
for all days of the week, showing the Highest Values (15.59 for S and P
CNX Nifty, 14.21 for S and P CNX 500) on Monday for the sample indices during
the study period. It indicates that the Return Distribution was not normally
distributed during the study period 
• 
The analysis of KruskallWallis Statistics shows that the
Test Statistic value was lower than the Table value (9.488) at 5% level
of significance in 4 degrees of freedom for the selected Index returns.
It clearly indicates that there was no significant difference between the
returns of different days of the week 
• 
The value of coefficients on Thursday was high and none of
the variables was statistically significant at conventional level of risk
in S and P CNX Nifty Index Returns. According to S and P CNX 500 Index Returns,
the highest value of coefficients was found on Thursday and it was statistically
significant at 5% level. But the insignificant Fvalue did not confirm the
Day of the Week Effect in the Indian Stock Market during the study period 
CONCLUSION
This study analyzed the Monday Effect for S and P CNX Nifty and S and P CNX 500 Index Returns. The study used the logarithmic data for sample indices in NSE and applied the Dummy Variable Regression Model. The result of the study found that there was the Highest Mean Return earned on Friday and the Lowest Mean Return earned on Monday for sample indices. The Seasonality Results indicate that there were no significant Days of the Week Effect in the Indian Stock Market during the study period. The study further reveals that Monday recorded the lowest returns it was the best period to buy the scrips (buy low). Friday shows high returns and it is the best period to sell the securities (sell high). The findings challenge the basic premises of the Efficient Market Hypothesis in its weakform ands this phenomenon could be considered as a superior opportunity for the investors to earn reasonable returns from the market.