AN ANALYTICAL STUDY ON SEASONAL ANOMALIES OF TEN (10) SENSEX (BSE) LISTED STOCKS FROM THE TIME PERIOD 2006 (FEBRUARY) TO 2014(FEBRUARY) Abstract G.Vignesh Prabhu Manager Placement & Sr. Lecturer, ISSM Business School, Chennai The twenty first century is the times were the world stock market focuses their view on the Indian Stock markets where there is a predominant growth prevails. India s focus towards the GDP of the country definitely makes the other Asian countries and the world economy tocus the investment in the Indian emerging Sectors. This paper attempts to study the dynamic Behavior of 10 Selected companies stock returns which are listed SENSEX (BSE ) from the time period 2006 (February ) to 2014(February), the study takes into consideration the companies who have turnover more than 500 Cr in a year. The selected companies are grouped into Five Emerging industries on the basis of industrial classification IT, Automotive, Metals & Mining, Banking and Pharmaceuticals each sector consists of Two top companies, in that particular industry which are listed in SENSEX from the year 2006 onwards. The primary objective of the study is to analyze the behavior of the expected stock returns in the selected 10 Indian companies, to analyze the independence of stock returns, to study the existence of Seasonal anomalies in stock returns, to analyze the share value from the specified period of the study. Using the index data s on the basis of daily movement the study will find the expected stock returns the favorable and unfavorable growth movement in the particular stock or the industry on a whole from the specified time period. The study is based on Secondary data to be collected from data base of (CMIE) Centre for Monitoring Indian Economy and (BSE) Bombay Stock Exchange website. The Analytical tools planned are average method, Frequency distribution Analysis, Auto correlation, further to this analytical tools and other statistical tools can also be added according to the requirements. Keywords: Stock Market, SENSEX, Turnover 144
Introduction This paper is a attempt to study whether there exists any seasonal anomalies in the Indian stock market, the stock market is a place were a lot of investment research aspects are undertaken, from the investors point of view they see the efficiency of the market conditions in order to make profits and to hold their stocks for future returns. Increased use of modern information technology and recent reforms in the operating mechanism of financial markets, have resulted in enhancing market efficiency. For instance Fama (1965); Jensen and Benington(1970); Rosenberg and Rudd (1982); French and Roll (1986); Mohanty (2002); and Bodla (2005) etc these are some of the researches undertaken in identifying seasonal anomalies in the stock market. Despite frequent claims with reference to market efficiency, literature on the subject shows numerous research works which offer evidence of seasonal/ calendar anomalies both in developed and emerging stock markets. Rozeff and Kinney (1976), were the first to document evidence of anomalies in NYSE stocks. They found evidence of high mean returns in January as compared to other months. French (1980), analyzed daily returns of stocks for the period 1953-1977 which showed a tendency for returns to be negative on Mondays but positive on other days of the week. Aggarwal and Tandon (1994), found significantly negative returns on Monday in nine countries and on Tuesday in eight countries out of the 18 countries taken for his study. A number of studies relating to stock market anomalies have been carried out in India also. But the results are still contradictory about emerging markets. To quote a few, I M Pandey (2002), concluded that the monthly effects exist in Indian stock market and investor can time their share investments to earn abnormal returns. Kiran Rotkar, Rishikesh Patel & Ashvin Patil (2002), using the data from January 1995 to December 1999 concluded that the stock returns are high on Wednesday and Monday while they are lowest on Friday. The study of Karmakar and Chakraborty (2003), indicates the presence of Friday effect, the monthly effect, the turn of the month effect and holiday effect in Indian stock market. All these steps have led to a spectacular growth of the markets in terms of the market capitalization, turnover and number of deals. This paper has been divided intour sections. First section gives introduction and review of 145
previous studies. The second section describes the data and methodology used to investigate the seasonal anomalies. While the third section contains the results of the study, final section gives the conclusions of the study. Data & Methodology Indian stock market is the best market to analyze the seasonal effects in a stock market the Indian investors tend to have many psychological bounds like day sentiments according to them they feel there are sentimental days auspicious days and they also have favorite days India is a fit case for this study as it is a fast growing market and has more depth in comparison. For achieving the objective of bringing out seasonal anomalies in India s stock market returns, SENSEX (a broad based index of BSE Ltd.) has been taken as a proxy to its market. The selected companies are based on convenience sampling method. SENSEX comprises of thirty most liquid individual stocks listed at Bombay Stock Exchange Ltd. (BSE). It is also considered as an indicator of the performance of whole economy. In this study we had taken 10 companies of BSE (SENSEX) and they are grouped industry wise the closing prices of all the 10 companies were collected from the websites of BSE and Yahoor the period ranging from February 2006 to February 2014. In order to avoid the influences of extreme index values the stock returns has been measured in terms of the continuously compounded daily percentage change in the concerned share price index,. Symbolically, Where Rt is the return in the period t; Pt is the daily closing share price index of a market at a particular time t;pt-1 is the closing share price index for the preceding period; In is natural logarithm. This paper assumes that there are four types of seasonal anomalies. These include (i) Friday effect, (ii) Monday effect, (iii) Semi- Monthly effect, and (iv) Monthly effect. To identify monthly effect, the mean daily return of last trading day of the month and the first three trading days of the month has been compared with the mean return for the rest of the days in the month. In the case of the semi-monthly effect, the mean return of the first half month (i.e. return on the 30th,31st calendar days of the preceding month and 1 to 14 th calendars days of the current month) are compared with the average return of the rest of the days. In order to analyze the Monday effect, the mean return of 146
Monday of the each week is compared with the average return of rest of the days. Similarly, return of the Friday is compared with the mean return of the rest of the days to identify the Friday effect. The significance of the difference between average returns was verified with the help of t-test by stating the following hypothesis: H0 : 1 = 2 H1 : 1 2 Where H0 is null hypothesis which state that there is no difference between the return of the first period and the second period, H1 is alternate hypothesis, 1 is the mean return of first segment and 2 is the mean of second segment. The level of significance is taken at 5 percent at which the critical region is -1.96<t<1.96. The t- test has been applied by using the following formulae. Where S2 p is pooled variance, n1 is number of observations in population 1 and n2 is number of observations in population 2, (μ1-μ2) is the difference between two population means and (X1- X2) is the difference between sample means. Analysis of variance is used to test the hypothesis that several means are equal. This technique is an extension of the two-sample t test. For the one to one comparison between months, Post- Hoc Test is used. The post-hoc test examines the difference between each pair of means, and yield a matrix where asterisks indicate significantly different group means at an alpha level of.05. Results of data analysis The results obtained from the analysis of data regarding the existence of seasonal anomalies in BSE ( SENSEX ) selected 10 stocks are presented in the table format from table 1 to 4 which is added in the appendix, Here table 3 is the mean analysis of all the 10 companies data s which were taken for the study. Table 3 Showing value of day effects S.No Friday Rest of days Monday Rest of days Semi Monthly Rest of days Monthly Rest of days 1 1.1363 0.7294 2.6026 0.6419 0.9654 0.9427 2.5296 0.685 From the above table it is clear that the mean value of all the attempted seasonal anamolies shows a higher return when compared to semi-monthly effect alone even though the semimonthly mean return has some growth comparatively its not good when we consider the other 147
effects like Friday, Monday and Monthly effects. Each anamolies are calculated separately and the results are give in table format which is added in appendix the results of the concerned tables are given below Table 3.1 Shows the results of Friday effect, Results obtained from the Friday return value effects are that the mean value, standard deviation, and variance of the all 10 companies are greater than the rest of days which indicates that the Friday returns are more than any other days in a month.. Table 3.2 Shows the results of Monday effect, Results - In the comparison between Monday and rest of days return s effect it is stated that the Monday mean value is higher than the other days mean value which shows that the Monday returns are more than other days and it is expected that in Monday the returns are more and trading is happening actively. Table 3.3 Shows the results of Semi-Monthly effect, Results shows that the returns in semi monthly and rest of days is differing company to company which shows that the results in semi- monthly can t be predicted which was earlier shown in Monday and Friday effects. Table 4.4 shows the results of Monthly effect. The monthly effect results shows that there is a definite growth in the first and last days of the month which results in the monthly effect anomaly in the Indian market Conclusions This study has been undertaken to examine whether there exist any seasonal anomalies in selected 10 Indian companies listed in BSE ( SENSEX). From the findings, thus it is obvious that some kind of seasonal anomalies like Friday effect, Monday effect and Monthly effect are persistent in the markets. Hence, despite the use of sophisticated information technology and after introducing many reforms, the stock markets are not fully efficient. The policy implications of the findings are as follows: The existence of anomalies may provide opportunities to formulate profitable trading strategies so as to earn the increased return that is not commensurate with the risk. As turn of the month affects the markets, investors can gor a trading strategy of buying stocks in the second half of the month and selling during the first half 148
of the month. The study shows that the return of the Monday has been lower in comparison to the return of the rest five days of the week. Accordingly, the investors might purchase the securities on Monday and sell them on other days. The above strategy would improve the performance of portfolios maintained by both individuals and institutional investors. However, the usefulness of the strategies remains questionable as the transaction costs and information costs of operating in stock markets have not been considered in the paper. Moreover, if such anomalies persist today and investors formulate their trading strategies accordingly, this would result in profit making only in the short run. In the long run, countervailing arbitrage and forces of demand and supply will exploit the excess return leaving nuture scope for such anomalies and the same would pave the way to make the market efficient. Still, the above strategy may be helpful in altering the timing of already scheduled purchase and sales transactions in both the stock markets under study. Another implication of the study arises because the efficiency of the stock markets is closely related to the allocation of scarce capital resources. This paper helps new investors in making them to understand about the existence of seasonal anomalies in the Indian stock market. References 1. Agrawal, A. and K. Tandon (1994), Anomalies or illusions Evidence from stock markets in eighteen countries, Journal of International Money and Finance 13, p.p.83-106. 2. Balaban E (1995), Day - of- the- Week Effects: New Evidence From an Emerging Stock Market, Applied Economics Letters, 2, p.p.139-143. 3. Board, J.L. and Sutcliffe, C.M. (1988) The Weekend Effect in UK Stock Market Returns, Journal of Business, Finance & Accounting, 1988, 15, p.p.199-213. 4. Cross, F. (1973), The Behavior of Stock Prices on Fridays and Mondays, Financial Analysts Journal, November/December 1973, p.p.67-69. 5. Fama Eugene F (1965), Random Walks in Stock Market Prices, Financial Analyst Journal 21, No. 5, Sept-Oct. 1965, p.p. 55-59. 149
Appendix Table 3.1 Friday effect in companies Friday NF SD F SD- NF Var F Var- NF N F N-NF T dff/nf Critical region CIPLA 0.1393 0.0337 5.2303 2.6438 27.3561 6.9898 399 1612 1.9039 2009/11-1.96<t<1.96 DRREDDYS 0.0963 0.0892 6.2765 2.4634 39.3956 6.0685 399 1612 0.1283 2009/11-1.96<t<1.96 GAIL 0.1909 0.0531 4.7019 2.6682 22.1086 7.1194 399 1612 2.4907 2009/11-1.96<t<1.96 HEROMOTORS 0.2852 0.0773 4.0570 2.2950 16.4600 5.2674 399 1612 8.3884 2009/11-1.96<t<1.96 HINDALCO 0.1273 0.0295 6.8477 3.4698 46.891 12.0399 399 1612 1.7743 2009/11-1.96<t<1.96 ICICI 0.3470 0.0886 6.5674 3.3020 43.1310 10.9032 399 1612 4.6822 2009/11-1.96<t<1.96 INFY 0.2006 0.0577 4.8647 2.6778 23.6659 7.1707 399 1612 2.5741 2009/11-1.96<t<1.96 MAHINDRA -0.1121 0.0879 8.0562 3.1899 64.9030 10.1755 399 1612-3.6436 2009/11-1.96<t<1.96 SBI -0.032 0.0748 5.6983 2.8424 32.4710 8.0793 399 1612-1.9286 2009/11-1.96<t<1.96 WIPRO -0.1062 0.0497 5.6847 2.7854 32.3167 7.7589 399 1612-2.8149 2009/11-1.96<t<1.96 Table 3.2 Monday effect in companies M NM SD-M SD -NM Var M Var NM N M N- NM T test Df M/NM Critical region CIPLA 0.123922 0.0364 5.2107 2.6503 27.1520 7.0244 402 1608 1.5820 2008/2010-1.96<t<1.96 DRREDDYS 0.3571 0.0924 4.7647 2.5278 22.7029 6.3902 402 1608 4.7792 2008/2010-1.96<t<1.96 GAIL 0.1896 0.0499 5.0099 2.5641 25.0999 6.5749 402 1608 2.5239 2008/2010-1.96<t<1.96 HEROMOTOR 0.2999 0.0757 4.3385 2.2231 18.8230 4.9422 402 1608 4.0429 2008/2010-1.96<t<1.96 HINDALCO 0.1616 0.0296 7.6279 3.3457 58.1859 11.1938 402 1608 2.4083 2008/2010-1.96<t<1.96 ICICI 0.3605 0.0920 6.8439 3.3880 46.8393 11.4787 402 1608 4.8827 2008/2010-1.96<t<1.96 INFY 0.2187 0.0568 5.1984 2.6343 27.0240 6.9400 402 1608 2.9266 2008/2010-1.96<t<1.96 MAHINDRA 0.4093 0.0868 7.1979 3.1226 51.8103 9.7507 402 1608 5.8720 2008/2010-1.96<t<1.96 SBI 0.2988 0.0759 5.7560 2.8777 33.1322 8.2815 402 1608 4.0374 2008/2010-1.96<t<1.96 WIPRO 0.1832 0.0464 5.5113 2.7130 30.3745 7.3608 402 1608 2.4743 2008/2010-1.96<t<1.96 Table 3.3 Semi monthly effects SM - ROD STD SM SD- ROD Var- SM Var- ROD N N T test Df SM/ROD Critical region CIPLA 0.0500 0.0266 3.1755 3.0540 10.0839 9.3271 984 1025 0.5279 2007/2009-1.96<t<1.96 DRREDDYS 0.1354 0.1375 3.1576 2.9092 9.9707 8.4634 984 1025-0.0480 2007/2009-1.96<t<1.96 GAIL 0.0737 0.0759 3.1926 3.1099 10.1933 9.6715 984 1025-0.0489 2007/2009-1.96<t<1.96 HEROMOTORS 0.1236 0.1137 2.8425 2.7178 8.0801 7.3867 984 1025 0.2235 2007/2009-1.96<t<1.96 150
HINDALCO 0.0451 0.0470 4.3461 4.2735 18.8890 18.2628 984 1025-0.0425 2007/2009-1.96<t<1.96 ICICI 0.1295 0.1383 3.9244 4.1482 15.4009 17.2082 984 1025-0.1971 2007/2009-1.96<t<1.96 INFY 0.0811 0.0874 3.2215 3.1774 10.3784 10.0964 984 1025-0.1425 2007/2009-1.96<t<1.96 MAHINDRA 0.1349 0.1263 4.0094 3.7967 16.0757 14.4156 984 1025 0.1947 2007/2009-1.96<t<1.96 SBI 0.1255 0.1132 3.8026 3.4169 14.4601 11.6754 984 1025 0.2776 2007/2009-1.96<t<1.96 WIPRO 0.0666 0.0768 3.4186 3.3360 11.6873 11.1290 984 1025-0.2286 2007/2009-1.96<t<1.96 Table 3. 4 Monthly effects in companies - M - ROD SD -M SD-ROD Var-M VAr-ROD N N T test Df Critical region CIPLA 0.1398 0.0261 5.0444 2.6342 25.4462 6.9391 388 1622 2.0281 2008/2010-1.96<t<1.96 DRREDDYS 0.3830 0.0873 5.2195 2.4366 27.2436 5.9374 388 1622 5.2738 2008/2010-1.96<t<1.96 GAIL 0.1829 0.0486 4.7124 2.5732 22.2068 6.6217 388 1622 2.3928 2008/2010-1.96<t<1.96 HEROMOT ORS 0.3046 0.0751 4.3498 2.2279 18.9212 4.9639 388 1622 4.0830 2008/2010-1.96<t<1.96 HINDALCO 0.1270 0.0242 6.8487 3.4003 46.9051 11.5624 388 1622 1.8451 2008/2010-1.96<t<1.96 ICICI 0.3500 0.0865 6.4386 3.2881 41.4563 10.8121 388 1622 4.7217 2008/2010-1.96<t<1.96 INFY 0.2161 0.0555 5.0343 2.5737 25.3442 6.6240 388 1622 2.8646 2008/2010-1.96<t<1.96 MAHINDRA 0.3172 0.0855 5.5924 3.1365 31.2752 9.8379 388 1622 4.1399 2008/2010-1.96<t<1.96 SBI 0.3214 0.0717 6.0426 2.7255 36.5134 7.4284 388 1622 4.4659 2008/2010-1.96<t<1.96 WIPRO 0.1876 0.0480 5.3873 2.7044 29.0236 7.3141 388 1622 2.4924 2008/2010-1.96<t<1.96 151