Jour of Adv Research in Dynamical & Control Systems, May 2017

Special Issue on Recent Trends in Engineering and Managerial Excellence

Seasonality Effect in Indian Stock Market

with Reference to BSE SENSEX Index
Dr.J. Sudarvel, Assistant Professor, Department of Master of Business Administration, Sri Ganesh School of Business
Management, Salem, Tamilnadu, India. E-mail:j.sudarvel@gmail.com
Dr.P. Dhanu, Assistant Professor, Department of Master of Business Administration, Sri Ganesh School of Business
Management, Salem, Tamilnadu, India. E-mail:dhanupra@gmail.com
Dr.R. Velmurugan, Associate Professor, Department of Commerce, Karpagam University, Karpagam Academy of Higher
Education, Coimbatore, Tamilnadu, India. E-mail:drvelsngm@gmail.com
Abstract--- The study tries to investigates the existence of the seasonality anomalies in Indian Stock Market. The
study utilizes the Daily return data of the Bombay Stock Exchange’s Sensex Index for the period ranging from April
2016 to March 2017 for analysis. The collected secondary data are analyzed by applying Descriptive statistics,
Linear Regression and paired ‘t’ test. The results of the study confirm the existence of seasonality in stock returns in
India and prevalence of the anomalies in Indian Stock Market. The results of the study confirm that the Indian stock
market is inefficient. Hence, investors can time their share investments to improve returns.
Keywords--- Indian Stock Market Anomalies, Seasonality Effect, Day of the Week Effect, Semi Month Effect and
Turn of the Month Effect.

I. Introduction
Stock market plays a vital role for the economic development of a country, by properly channeling the funds for
productive purpose. The proposition that a well-regulated stock market extends significant economic services is now
widely accepted and recognized by various academicians. Stock market assists economy as well as individual investors by
mobilizing the scarce resources and allocation in those sectors, which employ them optimally. Stock market assists
individual investors by providing continuous market for securities. From economic point of view, a well-developed stock
market has been considered requisite for economic growth as well as improvements in country’s productivity. The
progress of a country can be judged by ascertaining the stock market indicators like liquidity, asset pricing and turnover. In
addition, by ensuring a free and fair trading of stocks and performance of pricing mechanism, by ensuring a suitable return
on investment will ensure viable investment opportunities, in the stock markets acts as a driving force for channeling
savings into profitable investment and hence, ensures an optimal allocation of capital. In recent years, there has been
greater distress in the midst of shareholders, portfolio managers and researchers concerning about the behaviour of stock
market prices. The investors are eager to earn a higher rate of return on their investments. Hence, in order to satisfy
investor’s expectations, the portfolio managers have to look at the stock market conditions keenly and accordingly advise
investors and to construct a sound portfolio. Our country is believed to be as one of the fastest up-and-coming markets in
the world with a well-established stock market with a long history of organized trading in securities. Over the last few
years, advanced technology and online based transactions have modernized the stock exchanges whereas, in terms of the
number of companies listed and total market capitalization, the Indian capital market is considered large relative to the
country’s stage of economic development.
Stock market efficiency is amongst the most researched areas in finance. Stock markets are considered efficient
informational. The weak form of market efficiency states that it is not possible to predict stock price and return
movements using past price data. Following Fama (1965; 1970), a huge number of studies were conducted to test
the efficient market hypothesis (EMH). These studies mostly have shown that stock prices behave randomly. More
recently, however, researchers have collected evidence opposing to the EMH. They have identified systematic
variations in the stock returns. The important anomalies include the Day of the week effect, Semi Monthly effect
and the seasonal effect. The existence of the seasonal effect denies the weak form of the EMH and implies market
inefficiency. In an inefficient market investor, would be able to earn abnormal profits, that is, returns that are not
appropriate with risk.

II. Review of Literature

Archana. S, Mohammed Safeer and Kevin. S (2014) in their study entitled “A Study on Market Anomalies in
Indian Stock Market” found that the Turn of the month effect are minimally visible but not statistically proven for

ISSN 1943-023X 6
Jour of Adv Research in Dynamical & Control Systems, May 2017
Special Issue on Recent Trends in Engineering and Managerial Excellence

the analyzed period. Due to their timely actions prices of stocks quickly adjust to the new information, and reflect all
the available information. So, no investor can beat the market by generating abnormal returns. Ashish Garg, B.S.
Bodla and Sangeeta Chhabra (2010) in their study entitled “Seasonal Anomalies in Stock Returns: A Study of
Developed and Emerging Markets” found that the semi month effect does exists both in Indian stock market as well
as US stock market. Semi-Monthly Effect, the mean return of the First Half Month has been compared with the
average return of the rest half of the month. Dragan Tevdovski, Martin Mihajlov and Igor Sazdovski (2012) in their
study entitled “The Day of the Week Effect in South Eastern Europe Stock Markets” proves that the day of the week
effect is found only in Croatian and Bulgarian Stock Market. The day of the week effect does not exist in Bosnia,
Herzegovina, Macedonia and Serbia. Elena Valentina Ilica and Drago Opreaa (2014) in their study entitled
“Seasonality in the Romanian stock market: the-day-of-the-week effect” confirmed that the day of the week effect is
present on the Romanian stock market due to the existence of seasonality in the risk-return relationship. Faryad
Hussain, Kashif Hamid, Rana ShahidImdad Akash and Majid Imdad Khan (2011) in their study entitled “Day of the
Week Effect and Stock Returns: (Evidence from Karachi Stock Exchange-Pakistan)” proves that day of the week
effect does exist in Karachi stock exchange – Pakistan. Returns on Tuesday are more volatile over other days.
Empirical results of the study show that the assumption of the “EMH” is violated in different days’ domain.
Sudarvel. J and Dr. R. Velmurugan (2016) in his study entitled “Day of the Week effect in Indian Stock Market with
Reference to NSE Nifty Index” proves that day of the week effect does exist in Indian Stock Market. Shyam Lal Dev
Pandey and Gopi Prachetas (2012) in their study entitled “Testing of Risk Anomalies in Indian Equity Market by
Using Monthly Average Risk & Return” proves that low volatile stocks offer higher average rate of return than high
volatile stocks, which proves the existence of inefficiency in Indian Stock Market.

III. Objective of the Study

To identify the existence of the Day of the Week Effect, Semi Month Effect and Turn of the Month Effect in
Indian Stock Market.

IV. Research Methodology

Data
This study is analytical in nature. The data required for the study are collected from daily closing prices of BSE
Sensex index.
(Source: http://www.bseindia.com/indices/IndexArchiveData.aspx)
Period of the Study
The present study covers the period ranging between 1st April 2016 and 31st March 2017.
Framework of Analysis
The collected data have been analyzed by making use of Descriptive statistics such as Mean, Standard Deviation,
Variance, Skewness, Kurtosis and Shapiro-Wilk test, OLS Regression and paired ‘t’ test.
Limitation
Considering the continuity of data, only BSE Sensex Index has been selected for the study. Hence, utmost care to
be exercised while generalizing the results.
Day of the Week Effect
The standard assumption in financial theory is that the distribution of stock returns is identical for all weekdays,
however, stock market does not function during Saturday and Sunday. This break provides the possibility of day-of-
the-week effect, i.e. some weekday is different from other weekdays in the stock market. If the day-of-the-week
effect exists, some investors can take an advantage from it to make arbitrage.
MONDAY TUESDAY WEDNESDAY THURSDAY FRIDAY
Mean 0.0849 0.1770 0.0699 0.0666 -0.0635
Median 0.0684 0.1476 0.0074 0.1651 -0.0334
Std. Deviation 0.76409 0.78620 0.76198 0.79171 0.74902
Variance 0.584 0.618 0.581 0.627 0.561
Skewness 0.138 0-.160 0.758 -0.330 -0.788
Kurtosis 0.145 1.341 1.030 0.158 2.900
Shapiro-Wilk 0.837 0.015 0.019 0.301 0.007
Source – Database collected from BSE web portal and computed.

ISSN 1943-023X 7
Jour of Adv Research in Dynamical & Control Systems, May 2017
Special Issue on Recent Trends in Engineering and Managerial Excellence

High mean returns were noticed during Tuesday (0.1770) and low mean returns were found during Friday (-
0.0635). While comparing to variance, high level of volatility was noticed on Thursday (0.627) and low level of
volatility was noticed on Friday (0.561).
Result of the Skewness test disclosed that positive value and it was noticed during Monday returns (0.138) and
Wednesday returns (0.758), which implies that most of the Monday and Wednesday returns were more than the
average returns. The other day’s returns were found negative skewness, which implies that most of the other day’s
returns were less than the average returns. The Kurtosis result of the BSE SENSEX index returns were found
platykurtic on all the days of the week. Since, Kurtosis values are less than 3, thus, it is inferred that the level of risk
associated with all the day’s returns of the BSE Sensex index was low, which means investors may obtain either low
level of profit or loss.
From the p value of Shapiro-Wilk test BSE SENSEX INDEX returns is less than 0.05, in the Tuesday returns,
Wednesday returns and Friday returns, it is clearly proved that the data are not normally distributed in the Tuesday,
Wednesday and Friday returns. As the calculated P value of the Shapiro-Wilk testis greater than 0.05, in the Monday
return and Thursday returns it is clearly proved that the data are normally distributed Monday and Thursday returns.
Hence, there is an anomaly partially exists inthe BSE SENSEX index returns.
To measure day of the week effect, OLS is employed. Friday return is introduced as dependent variable and the
rest of the day, namely Monday, Tuesday, Wednesday and Thursday are considered as independent variables. The
following table illustrates the determinants of Day of the week effect of BSE Sensex Index.
Ordinary Least Square Regression
Coefficient Std. Error t-ratio p-value
const -0.0302716 0.113804 -0.2660 0.79151
MONDAY 0.376341 0.154819 -2.4308 0.01931 **
TUESDAY -0.0849992 0.140616 -0.6045 0.54870
WEDNESDAY 0.0543639 0.145734 0.3730 0.71096
THURSDAY -0.0907652 0.146574 -0.6192 0.53902
R-squared 0.124845 Adjusted R-squared 0.043435
F 1.533539 P-value(F) 0.029491
The result of Ordinary Least Square Regression disclose that Monday return is found to be significant at 5 per
cent level. Variables which are significant are discussed below.
Monday
The regression coefficient indicates that Monday returns positively influences Friday returns. The value of the
regression coefficient indicates that a unit of increase in Monday return shall increase Friday return by 0.376 units.
Higher rate of return on Monday leads to a higher rate of return on Friday.
The value of R2 is found to be significant at five per cent level. This shows that the regression equation framed is
a good fit. Around 12.48 per cent of variation in Friday return is due to select variables.
Semi Month Effect
In Semi month effect the mean return of the First Half of the Month has been compared to the return of second
half of the month. The purpose of studying Semi-Month Effect is to find appropriate investment time during
fortnight month.
Mean Std. Deviation Variance Skewness Kurtosis Shapiro-Wilk
First Half 0.0289 0.23271 0.054 -1.735 4.476 0.025
Second Half 0.1041 0.11925 0.014 1.12 2.034 0.010
Source – Database collected from BSE web portal and computed.
High level of mean returns is noticed during the Second half of the month (0.1041) and low level mean return is
found during the First half of the month (0.0289). Comparing the variance, high level of volatility was noticed on
the First half of the month (0.054) and low level of volatility was noticed in Second half of the month (0.014).
Further, the analysis of Skewness test discloses that negative returns are noticed in the First half of the month
returns, which implies that First half of the month returns were less than the average returns. And the Second half of
the month returns found positive Skewness, which implies that Second half of the month returns were more than the
average returns. The Kurtosis of BSE Sensex index returns were founded leptokurtic in the first half of the month
(4.476). Since Kurtosis values are greater than 3, thus, it is inferred that the level of risk associated with the first half
of the months returns of the BSE Sensex index was High, which means investors may obtain either high level of
profit or loss. The Kurtosis results of the BSE Sensex index returns were found platykurtic in second half of the

ISSN 1943-023X 8
Jour of Adv Research in Dynamical & Control Systems, May 2017
Special Issue on Recent Trends in Engineering and Managerial Excellence

month (2.034) Since, Kurtosis values are less than 3, thus, it is inferred that the level of risk associated with the
Second half of the month returns of the BSE Sensex index was low, which means investors may obtain either low
level of profit or loss.
From the p value of Shapiro-Wilk test BSE SENSEX INDEX returns is less than 0.05, it is clearly proved that
the data are not normally distributed there is an anomaly in the BSE SENSEX INDEX returns.
Mean N Std. Deviation Paired t value P value
FIRST HALF 0.0289 12 0.23271 -0.976 .350
SECOND HALF 0.1041 12 0.11925
Source – Database collected from BSE web portal and computed.
As the calculated P value is greater than 0.05, it is inferred that BSE Sensex index returns does not differ
between the first half and the second half of the month.
Turn of the Month Effect
The trend of stock prices to increase during the last two days and the first three days of each month, is called
Turn of the Month Effect. For the purpose of this study, the average return on the last two trading days of the
preceding month, and the first three days of the subsequent month were computed and compared with the mean
return for the rest of the days in the subsequent month.
Mean Std. Deviation Variance Skewness Kurtosis Shapiro-Wilk
First Half -0.0043 0.43708 0.191 -0.075 -0.298 0.998
Rest of The Month 0.0831 0.20161 0.041 -0.057 -1.184 0.662
Source – Database collected from BSE web portal and computed.
High level of mean returns is noticed during the Rest of the month (0.0831) and low level mean return is found
during the First half of the month (-0.0043). Comparing the variance, high level of volatility was noticed on the First
half of the month (0.191) and low level of volatility was noticed in Second half of the month (0.041).
Result of the Skewness test disclosed that positive returns were noticed in the First half of the month and rest of
the month returns, which implies that most of the First half of the month and rest of the month returns were more
than the average returns. The Kurtosis results of the BSE Sensex index returns were found platykurtic in both the
First half of the month (-0.425) and the rest of the month (-1.110). Since, Kurtosis values are less than 3, thus, it is
inferred that the level of risk associated with the first half of the month and rest of the month returns of the BSE
Sensex index was low, which means investors may obtain either low level of profit or loss.
As the calculated P value of the Shapiro-Wilk test is greater than 0.05, it is clearly proved that the data are
normally distributed. Hence, anomaly does not exist in the BSE Sensex index
Mean N Std. Deviation Paired t value P value
FIRST HALF -0.0043 12 0.43708 -0.532 0.583
REST OF THE MONTH 0.0831 12 0.20161
Source – Database collected from BSE web portal and computed.
As the calculated P value is greater than 0.05, it is inferred that BSE Sensex index returns does not differ
between the first half and the rest of the month.

V. Suggestions
It is necessary for the Indian Investors to carefully study the publicly available information, because it plays a
major role in analyzing the Market Efficiency and changes in the market. From the analyses of the Day of the Week,
it is suggested that investors may buy the shares on Friday and sell them on Tuesday because they may get better
returns than on other days.
The study found that the highest mean returns were documented in the second half of the month return in the
Semi-Month effect. Hence, Indian investors are advised to buy scripts during the First half of the month and sell
them on a Second half of the month period which will give better returns.
The study found that the highest mean returns was noticed in the Rest of the month. Hence, Indian investors are
advised to buy scripts during the First half of the month and sell the shares during Rest of the month, which will give
better returns.
We suggest that the user of this report to do homework before taking investment decision not based on this
report. The above outcome is based on a statistical tool used for analyzing the anomalies in the BSE Sensex index.

ISSN 1943-023X 9
Jour of Adv Research in Dynamical & Control Systems, May 2017
Special Issue on Recent Trends in Engineering and Managerial Excellence

There are various other factors to be considered before investing. We also suggest that the same can be applied in
penny stocks or stocks in minimum numbers before going for trading in full.

VI. Conclusion
This paper observed the presence of Day of the week effect, Semi-month and Turn of the Month Effect for BSE
Sensex in the Indian Stock Market. The finding of the Study shows that the Turn of the Month Effect was not
prevails in the Indian Stock Market. Semi Month Effect exists in the Indian Stock Market. Day of the week effect
anomaly partially exists in the Indian Stock Market. By ascertaining the prevalence of anomalies, Indian stock
market cannot be treated as fully efficient till now. The existence of these anomalies may provide the profitable
trade opportunities to the investors.

References
[1] Safeer, M. and Kevin, S. A Study on Market Anomalies In Indian Stock Market. International Journal of
Business and Administration Research Review 1 (3) (2014) 128-137.
[2] Garg, A., Bodla, B.S. and Chhabra, S. Seasonal anomalies in stock returns: A study of developed and
emerging markets. IIMS Journal of Management Science 1 (2) (2010) 165-179.
[3] Tevdovski, D., Mihajlov, M. and Sazdovski, I. The day of the week effect in South Eastern Europe stock
markets. Annals-Economy Series 3 (3) (2012) 20-24.
[4] Ţilică, E.V. and Oprea, D. Seasonality in the Romanian Stock Market: The-day-of-the-Week
Effect. Procedia Economics and Finance, 2014, 704-710.
[5] Faryad Hussain, Kashif Hamid, Rana ShahidImdad Akash and Majid Imdad Khan. Day of the Week Effect
and Stock Returns: (Evidence from Karachi Stock Exchange-Pakistan). Far East Journal of Psychology
and Business 3 (1) 2011.
[6] Sudarvel, J., Dhanu, P., Velmurugan, R. and Dharmaraj, A. Day of The Week Effect In Indian IT Sector
With Special Reference To BSE IT Index. International Journal for Scientific Research and Development
3 (8) (2015) 706-708.
[7] Sudarvel, J. and Velmurugan, R. Day of the Week Effect in Indian Automobile Sector With Reference To
BSE Auto Index. International Journal for Scientific Research and Development 3 (12) (2016) 361-364.
[8] Pandey, S.L.D. and Prachetas, G.K. Testing of risk anomalies in Indian equity market by using monthly
average risk & return. Management Insight 8 (2) (2013).
[9] Sudarvel, J. and Velmurugan, R. Indian Stock Market Anomalies: A Literature Review. International
Journal of Advance Research in Computer Science and Management Studies 3 (2) (2015) 173-178

Websites
[1] http://web.williams.edu/Mathematics/sjmiller/public_html/341/handouts/Fama_RandomWalksStockPrices.
pdf
[2] http://BrownMath.com/stat/shape.htm
[3] http://www.bseindia.com/indices/IndexArchiveData.aspx
[4] http://thismatter.com/money/investments/market-anomalies.htm
[5] http://finance.wharton.upenn.edu/~keim/research/NewPalgraveAnomalies(May302006).pdf
[6] https://books.google.co.in/books?id=1x09AAAAIAAJ&pg=PA35&lpg=PA35&dq=The+term+anomaly+ca
n+be+traced+to+Kuhn+(1970).&source=bl&ots=eQltqeGJbw&sig=64jZaaVz9f1s7fXLUgfvDKLVSYg&h
l=en&sa=X&ved=0CB0Q6AEwAGoVChMIh8j_xcORxwIVxQmOCh3oMQjJ#v=onepage&q=The%20ter
m%20anomaly%20can%20be%20traced%20to%20Kuhn%20(1970).&f=false
[7] http://www.rasch.org/rmt/rmt62e.htm
[8] http://calendar-effects.behaviouralfinance.net/weekend-effect/
[9] http://www.investorwords.com/7261/weekend_effect.html

ISSN 1943-023X 10