Volatility in the Emerging Stock Markets in Central and Eastern Europe: Evidence on Croatia, Czech Republic, Hungary, Poland, Russia and Slovakia

This paper investigates the main features of stock market volatility in the emerging markets of European transition economies using daily indexes. Starting with the universe of all stock markets in the transition economies, we use the criterion of data availability to obtain a sample of six stock markets, namely the markets in Croatia, Czech Republic, Hungary, Poland, Russia and Slovakia. We apply ARIMA, the BDSL procedure and symmetric as well as asymmetric GARCH models to test for daily return volatility. The main findings are fourfold. First, in all the six markets, volatility exhibits significant conditional heteroskedasticity and non–linearity. Second, volatility seems to be of a persistent nature; however, no asymmetric volatility effects are found for most of the markets. Third, as measured by a GARCH–in–Mean model, volatility does not explain expected returns for any of the six markets. Although GARCH appears to be the most appropriate process in characterising volatility in these markets, the explanation provided by symmetric and asymmetric GARCH models is not significant enough for predicting future volatility. Fourth, while the evidence suggests that the martingale hypothesis can be significantly rejected for all the six markets, none of the markets shows the well–known day–of–the–week anomaly commonly reported in most stock markets.