Volatility of stock returns:An empirical analysis of the Nairobi Stock Exchange

Abstract

This study analyzes the volatility structure of stock returns in an emerging stock market (NSE) covering the period 2nd January 1992 to 30th June 2003. The study utilizes daily stock returns calculated as log (Pt/Pt-1) where Pt represents the value of the NSE -20-share index at time t. The study uses both symmetric and asymmetric ARCH type models in investigating clustering effects, risk-return trade off, and volatility persistency, predictability and leverage effects. The results are as follows. NSE equity returns show negative skewness, excess kurtosis and deviation from normality. Returns are predictable and therefore rejecting the weak form efficiency. Asymmetric test results indicate that conditional volatility is higher with negative shocks implying a leverage effect. Consistent with most previous studies, a positive and significant relationship is indicated between conditional volatility and the stock returns implying that investors are risk averse. The positive and highly significant ARCH coefficient implies volatility clustering. Persistence of conditional volatility as measured by the sum of alpha and beta is less than unity, an indication that it is stationary (mean reverting) and therefore not explosive. The predictability of the second moments is not a random walk but a martingale process. The ARCH-LM test indicates that the returns are generated by a stochastic process and not a chaotic process. The institutional reforms that have taken place in the bourse, such as the entry of foreign investors in 1995 and change of trading system in 2000 are not significant in explaining volatility of the stock returns although the days of the week reflect significant negative returns while volatility is positive and significant on Monday, Tuesday and Thursday.