Asymmetric Volatility of Daily Stock Market Returns: Further Evidence from Nigeria

Abstract

In this study, we considered the asymmetric return-volatility relationship in the Nigerian stock market using 3179 daily market returns data for the period from 14^th January 2003 to 31^st December 2015. Two asymmetric GARCH models are employed; namely the TGARCH of Zakoian (1994) and the Power GARCH model of Ding, Granger and Engle (1993). We also assume that errors follow student-t distribution. BFGS optimization method with Marquardt procedure is employed to maximize the loglikelihood function. The risk-return relationship is also examined using the GARCH-in-mean framework suggested by Engle, Lilien and Robins (1987). Consistent with the recent findings of Ezirim, Nnaji and Ezirim (2017), the results of both TGARCH and PGARCH models show that there is no significant asymmetric volatility in the Nigerian stock market. Thus, returns and volatility are symmetrically related. The results also show that despite the high volatile nature of the Nigerian stock market, investors are not sufficiently rewarded. We therefore, conclude that both TGARCH and PGARCH models are not significantly different from the standard GARCH models in describing the time-varying properties of daily market returns in Nigeria.