Forecasting stock market volatility with non-linear GARCH models: a case for China

Abstract

This paper studies the performance of the GARCH model and two of its non-linear modifications to forecast China's weekly stock market volatility. The models are the Quadratic GARCH and the Glosten, Jagannathan and Runkle models which have been proposed to describe the often observed negative skewness in stock market indices. It is found that the QGARCH model is best when the estimation sample does not contain extreme observations such as the stock market crash, and that the GJR model cannot be recommended for forecasting.