Resumen
The changing patterns in returns on African stock markets have not been adequately documented. This paper addresses this gap by testing for the day-of-the-week effect and the changes in the patterns of returns for several African stock markets. A direct test on skewness and kurtosis is used to capture higher statistical moments in the search for seasonal patterns in returns. Daily index data for South Africa, Zambia, Botswana, Nigeria, and Morocco are used for the period 2004 to 2012. Day-of-the-week effects are documented for all the countries with the exception of South Africa. Furthermore, significant changes in patterns over time are observed for these same countries. Each day in the pre-financial crisis period shows significantly different patterns to every other day in post-crisis epoch. Further, the patterns displayed amongst the countries with significant results are largely similar in terms of highest/lowest mean returns. This paper presents new policy implications and research suggestions on the day-of-the-week patterns.