Professional Documents
Culture Documents
Seasonal Patterns in Indian Stock Markets: An Application of GARCH (1, 1) Model
Seasonal Patterns in Indian Stock Markets: An Application of GARCH (1, 1) Model
net
Research in Humanities, Arts
and Social Sciences
ISSN (Print): 2328-3734, ISSN (Online): 2328-3696, ISSN (CD-ROM): 2328-3688
AIJRHASS is a refereed, indexed, peer-reviewed, multidisciplinary and open access journal published by
International Association of Scientific Innovation and Research (IASIR), USA
(An Association Unifying the Sciences, Engineering, and Applied Research)
I. Introduction
The stock market and ups and downs of index or stock prices always gain the attraction from several sides. It is
the matter of wonder how stock prices move and hence several attempts have been made by many researchers to
estimate or forecast the stock prices. There are many methods of forecasting stock prices. Traditionally
fundamental analysis and technical analysis were used for this purpose. Fundamental analysis is related with the
intrinsic value of the stock. It involves analysis of company’s financial statements, profitability, capital
structure, dividend policy, industry specific factors etc. Technical analysts advocates that all financial and
market information is already reflected in the stock prices. Therefore, prices can only be predicted with the
relative forces of demand and supply. Thus it is very much related to price and volume data. As time passed,
researchers evolved many more new analysis techniques, one of them being calendar anomalies or seasonality in
stock markets. Seasonality is based on the concept that History repeats itself. It states that it is possible to make
excess or abnormal returns using past prices’ data. There are chances of earning excess returns at a particular
time-of-the-day, day-of-the-week, week or part-of-the-month, month-of-the-year, quarter-of-the-year, before or
after holidays etc.
Kaeppel (2009) says “Although things do change over time, but they tend to return to a particular state over and
over again.” For our purpose the term seasonality may be used to indicate any seasonal trends or cyclic and
repetitive fluctuations existing in stock market. Thus seasonality in stock markets has very little to do with
natural seasons i.e. stocks of only those industries directly or indirectly related to agricultural products are
having some connection with natural seasons. Apart these most stocks are affected by different other non-
natural seasons for example month-of-the-year season, day-of-the-week season, time of the day season, holiday
season, election cycle season, budget season, tax year end season etc. Each of the seasons is famous by other
adage such as The January Barometer, Sell in May and Go Away, Black Monday or Black Friday, November
effect etc.
Therefore, the present study proposes to explore existence and if existing, identify patterns of seasonality in
Indian stock market.
II. Review of available literature
Before proceeding towards data collection and analysis, a lot of earlier researches were reviewed and it was
found that seasonal behaviour of stock market has been studied using both non-parametric and econometric
techniques, but it was opined that econometric techniques can better study the seasonal behaviour as these
techniques are capable enough to handle problems like serial correlation and heteroskedasticity. Review of work
already done was classified on the basis of their objectives and findings as is presented in table 1.
It is concluded from Table 1 that most of the previous studies have focused on international markets and a very
few have been undertaken considering Indian stock markets. Most of the studies have used only closing prices
as if trading is done only at closing prices. Instead in the present study, average of daily open, low, high and
close prices have been used. Further, prior Indian studies had either SENSEX or Nifty as their samples and
sectoral indices were generally ignored. Therefore it was decided to take BSE indices including sectoral indices
as representative sample of the present study.
III. Research methodology
For the present study following research methodology has been adopted:
IV. Objectives
The study is proposed with the objective of exploring the following effects in the Indian stock market:
Day-of-the-week Effect
Month-of-the-year Effect
Monthly Effect
Quarterly Effect
V. Hypotheses
Accordingly, following hypotheses were formulated:
H01: Returns for all weekdays are equal.
H02: Returns for all months are equal.
H03: Returns for both halves of the month are equal.
H04: Returns for all quarters are equal.
VI. Data Collection
For studying the seasonal behaviour of stock prices, BSE was chosen as sample and its indices were included in
the study namely SENSEX, BSE -100, 200, 500, Midcap, Smlcap, Auto, Bankex, CD, CG, FMCG, HC, IT,
Metal, Oilgas, Power, PSU, Realty. Majority of prior researchers have used only closing prices as if trading is
done only at closing price. Rather average of these four prices can yield better results as it can control volatility
up to some extent. Therefore, daily close, open, high and low prices were collected for all indices from their
respective date of launching (or from the date when these four prices’ data were available) to 31 st May, 2013.
Thus very recent data have been collected. All these data have been collected from the official website of BSE
i.e. www.bseindia.com.
VII. Tools and Techniques
Following tools and techniques were used in the study:
Stationarity: Graphical analysis, Correlogram and ADF Unit Root Test
Serial Correlation:
o Detection: Serial Correlation LM Test
o Removal: ARIMA Modeling (Auto Regressive Integrated Moving Average)
Heteroskedasticity:
o Detection: ARCH-LM Test
o Removal: GARCH (1, 1) Model
IX. Software
Data preparation was done in MS-Excel while econometric analysis were performed in Eviews 7.0.
X. Stationary Analysis
First of all average prices were plotted on the line graph which showed a continuously changing patterns of the
prices (a sample has been shown for SENSEX in Figure 1). It gives a clue of non-stationarity. Then a
correlogram of the original average series for all indices was obtained (as sample Correlogram for SENSEX has
been shown in Table 2). It was observed from the correlogram that ACF did not declined even after increasing
the lag length showing non stationarity in the series. Next, Augmented Dickey-Fuller test was applied to
formally confirm the non-stationarity. The results of the test have been presented in Table 4. As it is clear that
all the test statistics are more than their critical values, the null hypothesis of stationarity is rejected. After
confirming non-stationarity, log differences of the all the series have been calculated using the following
formula:
r = [ ln ( Yt) – ln ( Yt-1 ) ] * 100
Where r is return, ln is natural log, Y t is absolute value of Y at time period t and Yt-1 is absolute value of Y at
time period t-1. After calculating returns, all the three tests were again applied. Graph of return series is
presented in Figure 2 which is showing mean reversion of the series. Correlogram as shown in Table 3 shows
that ACF immediately declines as soon as the lag length is increased. Further, test statistic of unit root ADF test
also confirms stationarity of the return series.
AVERAGE
24,000
20,000
16,000
12,000
8,000
4,000
0
92 94 96 98 00 02 04 06 08 10 12
Figure 1: SENSEX average series
RETURN
20
15
10
-5
-10
-15
92 94 96 98 00 02 04 06 08 10 12
Autocorrelatio Partial
n Correlation AC PAC Q-Stat Prob
Autocorrelatio Partial
n Correlation AC PAC Q-Stat Prob
Table 4: Results of ADF test* for SENSEX average and return series
t-Value t-Value
(Average Series) (Return Series)
Index
Intercept Intercept None Intercept Intercept None
and Trend and Trend
*Critical values of ADF test at 5% level of significance are -2.861890, -3.410652 and -1.940884 for intercept,
intercept & trend and none respectively.
XI. Dummy variable regression model
For the study of Day-of-the-Week effect, month-of-the-year effect, quarterly effect and monthly effect four
regression equations were formed including a constant term as Monday, January, First Quarter and First-half of
the month respectively. For remaining days, months etc. dummy variables were included in the equation.
Day-of-the-week
Yt = α1 + α2DTue + α3DWed + α4DThu + α5DFri + ε1 Equation 1
Month-of-the-year
Yt = α1 + α2DFeb + α3DMar + α4DApr + α5DMay + α6DJun + α7DJul + α8DAug + α9DSep +
α10DOct + α11DNov + α12DDec + ε1 Equation 2
Quarterly Effects
Yt = α1 + α2DII + α3DIII + α4DIV + ε1 Equation 3
Monthly Effects
Yt = α1 + α2DII + ε1 Equation 4
ε1 is the white noise error term.
XII. Results and discussion
To begin with, all the four equations were estimated for all BSE indices and results were not much exciting as
far as seasonality is concerned. There were little signs of calendar anomalies for a few indices. Other statistics
were very discouraging as the Adjusted R-squared and F- Statistic were not up to the mark, which indicated
poor fit of the model. The D-W statistic was also less than 2 for all the indices, which indicated serial correlation
in the return series. L-M test also confirmed this fact. To overcome this problem, proper ARMA terms were
added in the equation using Box-Jenkins Methodology and then the equation was estimated again for all the
indices. This time results got changed from those which were received earlier. The D-W statistic now became
very close to 2 and favorable results of LM-test, applied after inclusion of ARMA terms, confirmed the removal
of serial correlation. AIC and SBC statistics also got improved from the previous estimates.
Next, for checking existence of heteroskedasticity, ARCH-LM test was applied on the regression results
obtained after inclusion of ARMA terms. The significant test statistic confirmed the existence of
heteroskedasticity in all the indices of BSE. Therefore, GARCH (p, q) model with appropriate values of p and q
was estimated. ARCH-LM test was again applied on results generated by GARCH model and it was confirmed
that heteroskedasticity has been removed. Tables 5 to 8 present the estimation results of GARCH model for BSE
indices for different effects. The coefficients of different months have been presented with their respective z-
value in parenthesis. Significance of coefficients at 1%, 5% and 10% level has been indicated by *, ** and ***
respectively.
A. Day-of-the-week effect
Table 5 shows the results of GARCH (p, q) model for day-of-the-week effect. Surprisingly, similar to SENSEX,
Monday has been emerged to be significant for all indices of BSE except for BSE Power and Realty. Returns
for Monday are significant at 1% level of significance with the exception of being significant for FMCG and
Oilgas at 5% level and for IT and Metal at 10% level. Returns for Tuesday are significant only for 5 indices;
those for Wednesday are significant for 3 indices, for Thursday are significant for 4 indices and for Friday are
significant for 3 indices. With regard to sectoral indices, most of the indices are having the day-of-the-week
effect on Monday, only BSE CD (Consumer Durables) and CG (Capital Goods) are showing significant returns
on days other than Monday. Whereas other indices are having Tuesday and Thursday also as significant returns
generating days.
B. Month-of-the-year effect
Table 6 contains the results of GARCH (p, q) model for month-of-the-year effect and shows that there is major
transition in results, once heteroskedasticity is taken care of. SENSEX is displaying significant results in June,
September, November and December. Apart this, 12 out of 18 indices of BSE are showing significant returns in
September. BSE Realty did not show significant returns in any of the months. April effect is observable for only
two indices i.e. BSE CD and IT. Returns for February, March, July and October are not significant for any of the
indices. May effect is also not apparent in BSE indices as it is found to be significant only for BSE CD, CG and
HC and surprisingly, January effect was found only in case of BSE PSU.
Table 5: Results of GARCH model in BSE indices return series for day-of-the-week effect
Indices
BSE BSE BSE
Variables SENSEX BSE 100 BSE 200 BSE 500 BSE Auto BSE CD
Midcap Smlcap Bankex
0.138 0.157 0.229 0.202 0.197 0.166 0.161 0.193 0.247
C
(4.899)* (5.655)* (7.632)* (6.049)* (4.522)* (3.520)* (3.200)* (3.596)* (5.314)*
-0.0943 -0.104
-0.117 -0.129 -0.074 -0.041 -0.097 -0.041 -0.133
Tuesday (- (-
(-3.028)* (-3.026)* (-1.512) (-0.858) (-1.560) (-0.593) (-2.492)**
2.495)** 2.436)**
-0.085
-0.0244 -0.062 -0.058 -0.061 -0.039 -0.050 -0.038 -0.157
Wednesday (-
(-0.575) (-1.512) (-1.215) (-1.100) (-0.734) (-0.727) (-0.491) (-2.429)**
1.906)***
-0.082
-0.0514 -0.136 -0.082 -0.086 -0.046 -0.033 -0.041 -0.131
Thursday (-
(-1.243) (-3.110)* (-1.742) (-1.617) (-0.923) (-0.468) (-0.609) (-2.010)**
2.068)**
-0.094
-0.0112 -0.031 -0.104 -0.053 -0.064 -0.018 -0.051 -0.213
Friday (-
(-0.325) (-0.929) (-2.846)* (-1.418) (-1.667) (-0.330) (-0.798) (-4.350)*
2.356)**
Adj. R2 0.054473 0.078999 0.092983 0.096748 0.192032 0.291086 0.146389 0.117145 0.190662
AIC 3.263883 3.205853 3.115663 3.016866 2.792551 2.78441 3.069747 3.559979 3.346802
SIC 3.277386 3.219355 3.134119 3.040133 2.825978 2.820668 3.098542 3.588263 3.374342
Table 5 Contd.
Indices
Variables BSE BSE BSE BSE
BSE CG BSE HC BSE IT BSE PSU BSE Realty
FMCG Metal Oilgas Power
0.263 0.068 0.138 0.114 0.114 0.124 -0.013 0.138 -0.018
C
(5.298)* (2.165)** (4.913)* (1.898)*** (1.816)*** (2.569)** (-0.228) (3.694)* (-0.171)
-0.229 -0.002 -0.444 0.002 -0.041 -0.043 0.024 -0.064 0.081
Tuesday
(-3.524)* (-0.062) (-0.298) (0.034) (-0.510) (-0.691) (0.333) (-1.349) (0.659)
-0.176
0.026 -0.011 0.060 0.003 -0.024 0.092 -0.047 0.170
Wednesday (-
(0.565) (-0.277) (0.683) (0.038) (-0.333) (1.183) (-0.849) (1.249)
2.298)**
-0.16
0.012 -0.444 -0.001 -0.075 -0.061 -0.067 -0.091 0.011
Thursday (-
(0.257) (-1.088) (-0.010) (-0.857) (-0.899) (-0.867) (-1.619) (0.084)
2.185)**
-0.078
-0.073 -0.0002 -0.048 -0.079 -0.052 -0.060 -0.028 -0.054
Friday (-
(-1.234) (-0.007) (-0.788) (-1.100) (-0.970) (-0.934) (-0.693) (-0.486)
2.460)**
Adj. R2 0.055349 0.110712 0.082904 0.101917 0.116590 0.085267 0.069861 0.145122 0.113639
AIC 3.606070 2.762440 2.614313 3.836726 3.716085 3.210416 3.230276 3.074931 4.318741
SIC 3.622577 2.786296 2.634493 3.856907 3.744890 3.236593 3.272234 3.097090 4.358680
Table 6: Results of GARCH model in BSE indices return series for month-of-the-year effect
Indices
Variables
BSE SENSEX BSE 100 BSE 200 BSE 500 BSE Midcap BSE Smlcap
0.013361 0.032296 0.059698 0.091129 0.090311 0.108202
January (C)
(0.249224) (0.580825) (1.012631) (1.391073) (0.916245) (1.007286)
Table 6 Contd.
Indices
Variables
BSE Auto BSE Bankex BSE CD BSE CG BSE FMCG BSE HC
-0.039247 0.079097 -0.076665 0.021357 -0.031944 -0.012907
January (C)
(-0.412543) (0.815068) (-0.78628) (0.222230) (-0.524452) (-0.202944)
0.076782 -0.164631 0.098475 0.104389 0.021347 0.032578
February
(0.543916) (-1.179086) (0.76451) (0.767604) (0.232313) (0.373454)
0.100697 0.064637 0.037305 -0.088049 0.098935 0.125138
March
(0.700926) (0.407395) (0.26455) (-0.603592) (1.128065) (1.417906)
0.167254 -0.008125 0.264383 0.167078 0.131474 0.138327
April
(1.160681) (-0.053003) (1.8039)*** (1.147809) (1.342627) (1.586185)
0.030710 -0.033937 0.345312 0.358046 0.107135 0.205146
May
(0.237586) (-0.214174) (2.4345)** (2.682258)* (1.222569) (2.51691)**
0.072233 0.081461 0.236348 0.240996 0.204457 0.091560
June
(0.467669) (0.476111) (1.50929) (1.64935)*** (2.0000)** (0.942870)
0.135800 0.185328 0.114532 0.089191 0.124767 0.034053
July
(0.960337) (1.207944) (0.76852) (0.610621) (1.288173) (0.406652)
0.246676 -0.073507 0.246524 0.123870 0.137642 0.109894
August
(1.7298)*** (-0.486342) (1.672)*** (0.831802) (1.451798) (1.187222)
0.433771 0.495533 0.208002 0.189099 0.163953 0.215543
September
(3.162620)* (3.522732)* (1.400032) (1.359717) (1.8800)*** (2.4651)**
0.032270 -0.071010 0.170026 0.081963 -0.002415 0.039179
October
(0.242632) (-0.465498) (1.244938) (0.600148) (-0.025029) (0.443191)
0.261648 0.269448 0.501257 0.486690 0.173425 0.221460
November
(1.7229)*** (1.7303)*** (3.5621)* (4.060358)* (1.7786)*** (2.37868)**
0.245719 0.104444 0.245126 0.215880 0.150160 0.207032
December
(1.7236)*** (0.656807) (1.754)*** (1.443839) (1.605935) (2.25906)**
Adj. R2 0.145275 0.116936 0.188108 0.044672 0.111530 0.080314
Table 6 Contd.
Indices
Variables
BSE IT BSE Metal BSE Oilgas BSE Power BSE PSU BSE Realty
0.042604 0.046674 0.055188 -0.140866 0.137123 0.118623
January (C)
(0.526239) (0.409750) (0.52754) (-1.491718) (1.872)*** (0.563523)
0.087835 -0.179268 -0.095002 -0.050710 -0.128233 -0.304187
February
(0.646553) (-0.996408) (-0.65473) (-0.296564) (-1.15714) (-0.95563)
-0.016307 0.082141 0.043939 0.160791 -0.126868 -0.200956
March
(-0.12396) (0.455476) (0.29826) (0.969735) (-1.14747) (-0.65809)
Table 7: Results of GARCH model in BSE indices return series for quarterly effects
Indices
BSE BSE BSE BSE
Variables BSE 100 BSE 200 BSE 500 BSE Auto BSE CD
SENSEX Midcap Smlcap Bankex
Adj. R2 0.052465 0.076383 0.089795 0.093959 0.193041 0.290800 0.146806 0.118575 0.189407
AIC 3.264401 3.206940 3.118379 3.016356 2.791042 2.784468 3.066634 3.557005 3.350131
SIC 3.276676 3.219215 3.135297 3.037833 2.821696 2.817909 3.092810 3.582932 3.375834
Table 7 Contd.
Indices
Variables BSE BSE
BSE CG BSE FMCG BSE HC BSE IT BSE Metal BSE Oilgas BSE PSU
Power Realty
0.184665 0.115148 0.131334 -0.008465 -0.005256 0.086894 0.013719 0.108974 -0.067460
Quarter I (C)
(2.889421)* (2.724587)* (3.71640)* (-0.135956) (-0.061789) (1.289676) (0.177507) (2.2028)** (-0.47680)
AIC 3.602183 2.759571 2.614773 3.813517 3.714813 3.207761 3.231884 3.075418 4.317928
SIC 3.618695 2.781586 2.633118 3.835532 3.741000 3.231311 3.270027 3.095562 4.354236
Table 8: Results of GARCH model in BSE indices return series for monthly effects
Indices
Variables BSE BSE BSE BSE
BSE 100 BSE 200 BSE 500 BSE Auto BSE CD
SENSEX Midcap Smlcap Bankex
0.146533 0.149307 0.192114 0.198483 0.224139 0.207783 0.167797 0.210874 0.173198
First-Half (C)
(6.11027)* (6.181348)* (7.09458)* (7.309795)* (5.20459)* (4.5904)* (3.90227)* (4.4196)* (4.2652)*
Adj. R2 0.053483 .078082 0.091658 0.095837 0.194755 0.292933 0.147496 0.118284 0.187807
AIC 3.263126 3.205329 3.115974 3.022347 2.780929 2.778540 3.067127 3.556871 3.351253
SIC 3.272946 3.215149 3.129816 3.040245 2.808797 2.806407 3.088068 3.578083 3.373285
Table 8 Contd.
Indices
Variable
BSE CG BSE FMCG BSE HC BSE IT BSE Metal BSE Oilgas BSE Power BSE PSU BSE Realty
0.170280 0.077386 0.148892 0.129308 0.119661 0.087879 0.036133 0.146857 0.192844
First-Half (C)
(3.79271)* (2.779814)* (6.13528)* (3.406453)* (2.13164)** (2.02056)** (0.682633) (0.0000)* (2.0175)**
-0.066995
-0.005644 -0.091341 -0.056991 -0.086105 -0.012499 -0.112032 -0.106503 -0.328873
Second-Half (-
(-0.154272) (-2.8761)* (-0.964126) (-1.176995) (0.8239) (-1.714819) (-2.3857)** (-2.7889)*
1.128379)
Adj. R2 0.056240 0.112928 0.081884 0.101920 0.117332 0.086578 0.069636 0.147210 0.115334
AIC 3.604724 2.758472 2.612671 3.833375 3.713655 3.208494 3.228041 3.072479 4.311406
SIC 3.617566 2.780499 2.627348 3.851721 3.734605 3.229443 3.262370 3.088595 4.340452
D. Monthly effects
Table 8 contains the estimation results of GARCH (p, q) model for monthly effects. It is quite evident from
Table 8 that all non-sectoral indices of BSE and BSE HC, PSU and Realty are having significant returns in both
halves of the month whereas all other sectoral indices are having returns of first-half as significant with the
exception of BSE Power which has significant returns in neither part of the month.
XIII. Conclusions
The present paper is an attempt to explore the existence of seasonality in Indian stock market. Four types of
seasonal effects namely day-of-the-week effect, month-of-the-year effect, quarterly effect and monthly effect
were explored and for this purpose daily close, open, high and low prices were collected from the date of
launching of a particular index to 31st May, 2013. All BSE indices were taken as sample. For analysis of stock
market data, techniques of time-series econometrics were used. Results confirmed the existence of all the four
calendar anomalies in BSE. After analyzing all the four calendar effects namely day-of-the-week effect, month-
of-the-year effect, quarterly effects and monthly effects, it was found that all the four effects are present in the
Indian stock market. Amongst all weekdays, Monday was found to be most significant day for most of the
indices of BSE. In case of month-of-the-year effect, September and December were providing significant
returns. First quarter was emerged as the most significant quarter, when quarterly effects were tested. In case of
monthly effects, returns of first-half of the month were found to be significant. Further, once ARMA terms for
ARIMA modeling, order of ARCH for heteroskedasticity and p and q terms for GARCH model have been
identified in analyzing day-of-the-week effect, in analyzing other calendar effects all these parameters were
either equal or very close to their first estimates in day-of-the-week effect. It implies that while using
econometric techniques, basis characteristics of the series remains the same no matter how data are grouped
(daily, monthly etc.)
References
[1]. Algidede P, (2008), “Month-of-the-year and pre-holiday seasonality in African stock markets”, Stirling Economics Discussion
Papers with number 2008-23, Retrieved from http://ideas.repec.org/p/stl/stledp/2008-23.html visited on July 07, 2011.
[2]. Arumugam A, and Soundararajan K (2013), “Stock market seasonality- Time varying volatility in the emerging Indian stock
market”, IOSR Journal of Business and Management (IOSR-JBM), 9 (6), 87 (103). Retrieved from www.iosrjournals.org/iosr-
jbm/papers/Vol9-issue6/N09687103.pdf visited on October 05, 2013.
[3]. Bhabra H S, Dhillon U S and Ramirez, G G (1999), “A November effect? Revisiting the tax-loss selling hypothesis”, Financial
Management, 28 (4), 5-15. Retrieved from http://www.jstor.org/stable/3666300 visited on November 11, 2011.
[4]. Bhattacharya K, Sarkar N and Mukhopadhyay D (2003), “Stability of the day of the week effect in return and in volatility at the
Indian capital market: A GARCH approach with proper mean specification”, Applied Financial Economics, 13 (8), 553-563.
Retrieved from http://www.tandfonline.com/doi/ abs/10.1080/0960310021000020924#.UkpmZ3_ Mu1s visited on August 12,
2011.
[5]. Chan M W L, Khanthavit A and Thomas H (1996), “Seasonality and cultural influences on four Asian stock markets”, Asia
Pacific Journal of Management, 13 (2), 1-24. Retrieved from http://link.springer.com/article/10.1007/BF01733814 visited on
April 29, 2011.
[6]. Coutts J A and Sheikh M A (2002), “The Anomalies that aren't there: The weekend, january and pre-holiday effects on the all
gold index on the Johannesburg stock exchange 1987-1997”, Applied Financial Economics, 12 (12), 863-871. Retrieved from
http://www.tandfonline.com/doi/abs/10.1080/09603100110052172#.UkpybH_Mu1s visited on July 28, 2011.
[7]. Dash M, Dutta A and Sabharwal M (2011), “Seasonality and market crashes in Indian stock markets”, Asian Journal of Finance
& Accounting, 3 (1), 174-184. Retrieved from http://www.macrothink.org/journal/index.php/ajfa/article/ download/997/1046
visited on July 12, 2012.
[8]. Gao L and Kling G (2005), “Calendar Effects in Chinese Stock Market”, Annals of Economics And Finance, 6, 75-88. Retrieved
from http://www.aeconf.net/Articles/May2005 /aef060105.pdf visited on May 03, 2011.
[9]. Guo S and Wang Z (2007), “Market efficiency anomalies: A study of seasonality effect on the Chinese stock exchange”
(Doctoral Dissertation). Retrieved from http://www.diva-portal.org/smash/get/diva2:141436/FULLTEXT01.pdf visited on May
04, 2011.
[10]. Joshi N and Bahadur K C F (2005), “The Nepalese stock market: Efficiency and calendar anomalies”, Economic Review:
Occasiona Paper of Nepal Rastra Bank, 17, 43-87. Retrieved from http://mpra.ub.uni-muenchen.de/26999/ visited on March 26,
2011.
[11]. Kaeppel J (2009). Seasonal Stock Market Trends: A Definitive Guide to Calendar Based Stock Market Trading. John Wiley &
Sons, p.15-16.
[12]. Kuria M A and Riro G K (2013), “Stock market anomalies: A study of seasonal effects on average returns of Nairobi securities
exchange”, Research Journal of Finance and Accounting, 4 (7), 207-215. Retrieved from
http://webcache.googleusercontent.com/search?q=cache:esA5tkYoiwEJ:www.iiste.org/Journals/index.php/RJFA/article/downloa
d/6301/6662+&cd=21&hl=en&ct=clnk&gl=in&client=firefox-a visited on August 20, 2013.
[13]. Lakonishok J and Smidt S (1988), “Are seasonal anomalies real? A ninety-year perspective”, The Review of Financial Studies, 1
(4), 403-425. Retrieved from http ://umdrive.memphis.edu/cjiang/www/teaching/fir8-7710/paper/lakonishok_smidt_1988_
rfs.pdf visited on August 10, 2011.
[14]. Lean H H, Smyth R and Wong W-K (2007), “Revisiting calendar anomalies in Asian stock markets using a stochastic dominance
approach”, Journal of Multinational Financial Management, 17 (2), 125-141. Retrieved from http:// www.sciencedirect.com/
science/article/pii/S1042444X06000429 visited on August 22, 2011.
[15]. Nageshwari P and Selvam M (2011), “An Empirical study on seasonal analysis in the Indian stock market”, International Journal
of Management and Business Studies, 1 (4), 90-95. Retrieved from http://www.ijmbs.com/14/nageshwari.pdf visited on March
25, 2011.
[16]. Ogunc A, Nippani S and Washer K M (2009), “Seasonality tests on the Shanghai and Shenzhen stock exchanges: An empirical
analysis”, Applied Financial Economics, 19 (9), 681-692. Retrieved from http://www.tandfonline.com/doi/abs
/10.1080/09603100802167296#.Ukp0jn_Mu1s visited on June 17, 2011.
[17]. Pandey I M (2002). “Is there seasonality in SENSEX monthly returns?” IIMA Working Papers with number WP2002-09-08 by
Indian Institute of Management Ahmadabad, Research and Publication Department. Retrieved from
http://www.iimahd.ernet.in/publications/data/2002-09-08IMPandey.pdf visited on July 12, 2011.
[18]. Pathak M R (2013), “Stock market seasonality: A study of the Indian stock market (NSE)”, PARIPEX-Indian Journal of
Research, 2 (3), 200-202. Retrieved from http://theglobaljournals.com/paripex/file.php?val=OTY4 visited on June 15, 2013.
[19]. Ray, Sarbapriya (2012). “Investigating seasonal behavior in the monthly stock returns: Evidence from BSE SENSEX of India”,
Advances in Asian Social Sciences, 2 (4), 560-569. Retrieved from http://webcache.googleusercontent.com/search?q=cache:
ZJ_u10V_WcgJ:worldsciencepublisher.org/journals/index.php/AASS/article/view/642/539+&cd=29&hl=en&ct=clnk&gl=in&cli
ent=firefox-a visited on March 21, 2013.
[20]. Rompotis G G (2009), “A comprehensive study on the seasonality of Greek equity funds performance”, South-Eastern Europe
Journal of Economics, 2, 229-255. Retrieved from http://www.asecu.gr/Seeje/issue13/Rompotis.pdf visited at July 17, 2011.
[21]. Shakila B, Prakash P, and Babitha R (2013), “Anomalies in Indian stock market: evidence of the day-of-the-week effect with
reference to National Stock Exchange, India”, ZENITH International Journal of Business Economics and Management Research,
3 (7), 72-81. Retrieved from http://www.indianjournals.com/ijor.aspx?target =ijor:zijbemr&volume=3&issue=7&article=008
visited on October 5, 2013.
[22]. Yakob N A, Beal D and Delpachitra S (2005), “Seasonality in the Asia Pacific stock market”, Journal of Asset Management, 6
(4). 298-318. Retrieved from http://eprints.usq.edu.au/882/ visited on March 30, 2011.