Determining the Existence of Calendar Anomalies on Stock Returns at the Nairobi Securities Exchange Market- Pre and Post the 2015 Reforms

Abstract

Random movements of stock returns have been studied since the introduction of the Efficient Market Hypothesis in the 1960s. Particularly, a range of stock calendar anomalies such as the Monday effect, the incredible January effect, and holiday effects have been explored and documentary especially in developed capital markets. Though such studies have covered developing capital markets, mixed results have been generated regarding the existence of calendar anomalies and stock returns seasonality. At the Nairobi Security Exchange (NSE), many studies have only investigated the size effect and the holiday effects on stock returns. Differences in methodological approaches sample size used have been identified as possible reasons for the mixed results regarding calendar anomalies at the NSE. This study, therefore, sought to determine the existence of calendar anomalies at the NSE pre and post the 2015 structural, operational, and institutional reforms introduced by the capital markets authority (CMA). The study used panel data collected from the NSE covering average daily, weekly, and monthly returns from January 2010 to December 2017 and analyzed using STATA. The study was theorized on the Random Walk and Efficient Market hypotheses by Burton and Eugene Fama respectively. Markov’s Regime-Switch Models (MRS) particularly the Calendar Time Hypotheis (CTH) by Tang was the theoretical framework used in the study. Anderson-Darling test was conducted to determine the normality of the distribution. Another diagnostic test done was the Breusch-Pegan heteroscedasticity test. In examining the day-of-the-week effects, the study found that stock returns were lower on Mondays and Fridays both pre and post the 2015 reforms, these were statistically insignificant to justify the Monday and the day of the week effects. The January effect or month of the year effect was not observed at the NSE. Therefore, the study concluded that the 2015 reforms did not contribute significantly making the NSE more efficient as was intended by the CMA. Undoubtedly, the findings of this study will not only fill the empirical gaps but also be very helpful to stockbrokers and investors with passion for stock investments.