Volatility Forecasting with Asymmetric Normal Mixture Garch Model: Evidence from South Africa

This paper investigates the relative performance of the asymmetric normal mixture generalized autoregressive conditional heteroskedasticity (NM-GARCH) and the benchmarked GARCH models with the daily stock market returns of the Johannesburg Stock Exchange, South Africa. The predictive performance of the NMGARCH model is compared against a set of the GARCH models with the normal, the Student-t, and the skewed Student-t distributions. The empirical results show that the NM-GARCH outperforms all other competing models according to Christoffersen’s (1998) tail-loss and White’s (2000) reality check tests. This evidence shows that mixture of errors improves the predictive performance of volatility models.