Analyzing And Forecasting Chinese Stock Market Volatility Using Multifractal Theory

With the development of financial market reform and innovations,the prevention of financial risks and strengthen the financial risk management has become the primary issue of the development of international financial markets.Precise forecast of volatility of return on financial assets is an effective way to control financial market risk.Therefore,finding appropriate tools and methods for measuring and forecasting the volatility of financial assets is a top priority for financial risk management.The fractal market hypothesis can better explain the vision of financial market,and has important theoretical value and practical significance on major issues such as financial market risk management,market regulation and price forecasting.In this paper,MF-DMA is used to study to study the multifractality of Chinese stock market and its causes.The empirical results show that Chinese stock market does have the characteristics of multifractal,which further illustrates the applicability of multifractal theory to the study of China's stock market.The empirical results also show that fat-tail distribution and extreme events together lead to the multifractal characteristics of the Chinese stock market,while the intrinsic long-range correlation has little effect on it.This article also find that Chinese stock market is easier to be affected by information compared with the mature financial markets like American stock market.The uncertainty of Chinese stock market mainly comes from the extreme events such as government regulation.We accordingly put forward a series of policy suggestions in the hope of providing new research ideas for the volatility risk management in the Chinese stock market.In addition,this paper constructs a variety of ARFIMA-GARCH class models and compare their forecast accuracy of Chinese stock market volatility with the multifractal theory-based Markov switching multifractal volatility model(MSM).The empirical results show that the MSM model is better than the GARCH models both in the goodness of fit in the sample and the prediction accuracy outside the sample.The conclusion that the volatility of the Chinese stock market is predictable has very important practical significance.The research in this paper provides valuable empirical results for the financial risk management.