Empirical Analysis of Chinese Stock Market Volatility Based on GARCH Models and Markov Switching Models

Volatility has been the focus in the financial field in recent decades. It can be used to measure the uncertainties of yield and represent the risk of assets. In this paper, GARCH models and Markov switching models are used to fit the volatility of the Chinese stock market. Results illustrate that Markov switching models take the regime-switch as an endogenous variable and a random process, which enable it to describe all the remarkable structural change in one united model and help to forecast price. Therefore, it is superior to GARCH models.