Study On Best Choice Of GARCH Model With Different Assumptions And Its Influence On Volatility Spillover Effect In Stock Market

The stock market is the most concern is the volatility,volatility trend fluctuation determines the investment income,but investors return is the premise is to ensure the normal fluctuations in the stock market,which has a great influence and effect on the stock market norms and healthy development.The volatility of stock price is the most outstanding characteristic of the stock market,how to better characterize the volatility of stock price fluctuation to grasp the stock market rules,all investors are most concerned about the issue,but also become a hot issue discussed by many scholars.China’s stock market is still a relatively young market,is in the development and transition stage has its unique fluctuation characteristics,speculative,volatility is more intense,easy rose fell situation,the fluctuation of high frequency,large amplitude,and easy to influence by the policy oriented.At the same time,with the development of society,financial globalization and financial market opening,the”volatility spillover”between stock markets was playing even more important role in decision-making of investors.China’s securities market is also in rapid development,Shanghai and Shenzhen two regions of the stock market relevance is also increasing.Therefore,it is of great practical significance to strengthen the research on stock price volatility and the linkage with other markets in the current market environment.In order to study the volatility spillover effect on Chinese stock market,to inquire a large number of previous literature on the theoretical level,on the basis of the summary,with clear description of the causes and mechanism of stock market volatility spillover,the knowledge for the further research on stock market volatility spillover.In the empirical analysis,the residual distribution GARCH model this paper compares three different distribution hypothesis and select the best fitting on the stock market,using GARCH model and innovation after the residual sum of squares as the research object,the Grainger causality test,in order to explore the conduction between the volatility of China’s stock market.This paper selects the Shanghai Composite Index and Shenzhen component index as the research object,take the latest historical data of each index respectively,from November 8,2013 to December 31,2016 a complete stock market cycle(bull,bear period,consolidation period)the stock price data.Through the statistical analysis of their logarithmic yield data.First of all,we find that China’s stock has the characteristics of peak thick tail and volatility clustering,which does not conform to the assumption of normal distribution.It is proved that the returns of Shanghai Composite Index and Shenzhen component index are conditional heteroscedasticity,which is suitable for GARCH model.Therefore,we use the E-views software to establish three different residual distribution GARCH model based on China’s Shanghai and Shenzhen two stock stock market volatility,confirmed its fluctuation characteristics.And China’s stock market has a strong volatility persistence and long memory,once the impact of abnormal return of stock returns,it is difficult to eliminate in a short period of time.After that,this paper uses the AIC criterion and the prediction error index as the evaluation criteria,and examines the three kinds of different tail GARCH model to describe the volatility of China’s stock market and its volatility forecasting ability.For the comparison of model volatility forecasting ability,the different distribution of GARCH model in forecasting ability of the sample period and the sample period of the model.The sample was divided into November 8,2013 to October 31,2016 and from November 1,2016 to December 31,2016,the previous data estimation model was predicted,and the latter data were tested.Study of sample period is also selected forecast data of two months of November 8,2013 to December 31,2016 forecast,January 1,2017 to February 28,2017 data inspection.The results show that both the Shanghai Composite Index and Shenzhen component index,the GARCH-GED model is the best fitting model and the best model for forecasting volatility.Finally,select the optimal GARCH-GED of two shares of the logarithmic return rate of data modeling,Grainger causality test GARCH-GED model fitting residual data extraction,analysis of China’s stock market volatility spillover.The results show that between China’s Shanghai two stock market of the area,only the Shanghai stock market has significant volatility spillover effects on the Shenzhen stock market.This shows that the volatility of stock market is a one-way transmission of Shanghai stock market into the Shenzhen stock market.