An analysis of skewness and skewness persistence in three emerging markets

This paper reports an investigation into the extent and persistence of skewness in stock returns in three emerging markets, namely the Czech Republic, Kenya and Poland. The study is undertaken using the extended skew normal distribution and an asymmetric version of the generalised error distribution. The motivation for this paper is the hypothesis that skewness is a particular feature of returns in emerging markets; it may lack persistence and may decline in absolute terms as time passes and the market matures. When daily returns are considered, the majority of stocks in all three markets exhibit a significant degree of skewness. The value of the skewness parameter is often different in each of the three estimation periods considered. Little evidence has been found to support the view that skewness is an artifact of emerging or evolving markets. Over the period covered by the study, the number of stocks with a significant degree of skewness has remained more or less the same. For weekly returns, the same conclusions apply to the Czech Republic and to Kenya, but there is far less evidence of skewness in weekly returns on Polish Stocks. There is consistent evidence of short-term reversion in daily returns; increases (decreases) in mean return and volatility imply that there will be a decrease (increase) in skewness in the next month. This effect does not persist over longer time horizons.