An Investigation of Downside Risk in Equity Markets

The purpose of this study is to investigate the forecast power of the predicted volatilities for the downside risk in the Nikkei 225. Our empirical examinations clarify that the predicted volatilities both from the standard GARCH (Generalized Autoregressive Conditional Heteroskedasticity) model and the GARCH-in-mean model possess statistically significant forecast power for the extreme price drops in the Nikkei 225. Moreover, this paper also clarifies that the forecast power of the predicted volatilities by the GARCH-in-mean model is much stronger than that by the standard GARCH model.