Day-of-the-week and turn-of-the-month effects on stock market in Malaysia

Abstract

The existence of seasonality in stock returns have challenged the notion under the Efficient Market Hypothesis. This paper examines the existence of day-of-the-week and turn-of-the-month effects in the stock returns of the Kuala Lumpur Composite Index (KLCI) in Malaysia for the period from 2000 to 2019. Using the ARIMA-EGARCH model, persistency of both effects are investigated using four shorter sub-samples. The findings from the whole sample show that while the day-of-the week effect is only present on the volatility but not on the mean returns, Monday is found to have the lowest returns and the highest volatility. The day-of-the-week effect on the volatility is found to persistently present while the absence of the effect on the mean returns is observed in three of the subsamples. Although the turn-of-the-month effect is found to present in the whole sample and quite persistently in the sub-samples, the effect is absent in the recent years. Evidence from sub-samples shows that while investors may obtain higher mean returns during the month-end period, the month-end period also present higher volatility than that during the mid-month period. Further analysis revealed that Monday is not necessarily a bad day at the stock markets with returns on Monday that fall during the month-end period are generally found to be higher than those on other days of the week. The absence of certain effects during shorter sub-samples imply that investors need to revise their trading strategy and that any strategy is only beneficial for short-term period.