Analysis of market efficiency: The day of the week and the month of the year effect in ISE

The market anomalies were observed in several studies in contrast to Efficient Market Hypothesis. The day of week effect and the month of year effect are the anomalies that are investigated most in the stock markets. Depending on this fact, this study investigates the existence of the day of week effect and the month of year effect in Istanbul Stock Exchange (ISE) on the basis of ISE 100 Index returns over the period from 1995 to 2011. In this study, Kruskal Wallis test and GARCH models were used to analyze these effects in ISE.The returns on Tuesdays, Wednesdays and Thursdays are positive and statistically meaningful according to the results of the regression analysis with respect to the day of week effect over the period from 1995 to 2011. Furthermore, the returns on Wednesdays were found to be higher than the returns of other days. Similarly, the day of week effect was detected over the period between 1995 and 2007. The returns on Friday were found positive and statistically meaningful for the period 2008?2011. Next, downward trends in ISE over the period between 1995 and 2011 were added on model as a dummy variable. Consequently, the day of week effect become visible more clearly and the returns of all days of the week were found positive and statistically meaningful. Moreover, the highest average returns were observed on Wednesday.The results of the regression analysis of the month of year effect over the period between 1995 and 2011 indicate that the returns on March, April, July, September, October, and December are positive and statistically meaningful. Besides, the returns on October are higher than the returns of other months. Finally, depending on the observed anomalies, the results of all analysis show that ISE was not a weak form efficient.Keywords: Market Efficiency, Efficient Market Hypothesis, Market Anomalies, The Day of the Week Effect, The Month of the Year Effect