An Empirical Analysis of Calendar Anomalies in Stock Returns Evidence from India

Abstract

The key objectives of the present study are to investigate the existence of calendar anomalies in the Indian stock market. For this study, the following four indices, namely, SandP CNX Nifty, SandP CNX 500 Index, BSE Sensex and BSE 500 Index have been taken. The total period was divided in two short windows and long windows respectively from April 2002 to March 2008, April 2008 to March 2014 and April 2002 to March 2014. The OLS estimation indicates that there is no day of the week effect exists except Tuesday effect in sub-period II in any of the period of any indices except the SandP CNX Nifty Index. In the case of monthly effect, every window of all indices does not show any regular pattern. It was concluded that returns pattern are not similar in each study period but generally lowest in January and highest in December and volatility is also lower in mid of years. In case of semi-monthly effect, mean a daily return of first half is far better than second half in all windows for all indices. Volatility is more in second half of the month than the first half. In case of the turn of the month effect, it can be concluded that in all periods of the calendar months of all indices understudy, mean the daily return of turn half is far better than the mid half. OLS estimation in all indices indicates that there is a significant difference between returns of turn half and mid half of the calendar months. In the case of the Holiday Effect, during the whole period study of all indices undertaken, the mean return and volatility of weekdays is higher than pre-holidays and post-holidays trading days. The OLS estimation indicates that there is no Holiday effect exists in any of the periods under the study of all indices.