The Empirical Researches On Volatility Of Shanghai Stock Market Based On A Non-linear GARCH Models

In financial applications, the conventional GARCH model has arguably been the most popular model for conditional variance. Recently, however, growing evidence has suggested that the typically observed high persistence implied by this model does not characterize the behavior of stock returns. In particular, it has been demonstrated that the conventional GARCH model can exaggerate volatility persistence compared to the (true) volatility process perceived by the market. In the article,at first, we reviewed relevant documents about ARCH model, and introduced a new kind of nonlinear GARCH models. Special interest is devoted to models that are similar to previously introduced smooth transition GARCH models expect for the novel feature that a lagged value of conditional variance is used as the transition variable. This choice of the transition variable appears more suitable than the previous alternatives when the goal is to model series with highly persistent volatility. Firstly, this article finds out we can do empirical research on the asymmetries of the volatility of shanghai stock market, through GARCH model based on t-distribution. At last, an empirical application to shanghai stock return series demonstrates the differences between the new model and conventional GARCH models. The conclusion is that the estimation results of the nonlinear GARCH models can guaranteeing stability of the model. And we find that the conventional GARCH model can exaggerate volatility persistence, but the new nonlinear GARCH models can accurate characterize it. So we can better known the characters of our stock market, investors could evading risks, provide policy advice on decision making for the government, only to benefit our stock market's prosperity and development.