Modelling and forecasting volatility of the Botswana and Namibia stock market returns: evidence using GARCH models with different distribution densities

Abstract

This paper estimates and compares alternative distribution density forecast methodology of three generalised autoregressive conditional heteroscedasticity (GARCH) models for Botswana and Namibia stock market returns. The symmetric GARCH and asymmetric Glosten Jagannathan and Runkle (GJR) version of GARCH (GJR-GARCH) and exponential GARCH methodology are employed to investigate the effect of stock return volatility in both stock markets using Gaussian, Student-t and generalised error distribution densities. The evidence reveals that the current shocks to the conditional variance will have less impact on future volatility in both markets. News impact is asymmetric in both stock markets leading to the existence of leverage effect in stock returns. Besides, both markets exhibit reverse volatility asymmetry, contradicting the widely accepted theory of volatility asymmetry. Regarding forecasting evaluation, the results reveal that the symmetric GARCH model coupled with fatter-tail distributions present a better out-of-sample forecast for both stock markets.