International Stock Market High Frequency Volatility Information And China's Stock Market Volatility Forecasting Research

Financial globalization has become an important trend in the development of the world's financial system.In the context of accelerated integration of financial markets,the trend of mutual penetration and mutual influence of financial markets between countries or regions is increasingly evident.This effect is not only reflected in the return on assets,but also in the aspect of asset volatility,that is,financial market volatility is not only affected by its own previous volatility,but also by the volatility of other financial markets,which is called "the volatility spillover effect."With the rapid improvement of China's economic status in the world,especially with the transformation and opening up measures in the field of capital markets in the past 20 years,China has gradually integrated its stock market into the global stock market,and has made it increasingly connected with other stock markets.It means that the influence of international financial markets on China's financial market is intensified,the influence of international market fluctuation information is more extensive,and the transmission effect of fluctuation is more rapid.When risk regulators and international investors predict the Chinese stock market,they usually make decisions and trade with reference to the international stock market.Therefore,it is necessary to establish the overall concept of the global stock market,explore more suitable methods,analyze the characteristics of the impact of the international stock market on China's stock market,and better study the global stock market information to help managers and investors predict the volatility of the Chinese stock market.Measuring the impact and forecasting ability of the international financial market on the Chinese financial market is conducive to timely and accurate preventive measures in the face of international market risks,and is of great forward-looking significance for maintaining the security and stability of China's financial market.Although many scholars at home and abroad have studied the volatility of the financial market,so far,there are still many obvious methodological deficiencies that need to be resolved and improved how to accurately and fully use the information of international stock market volatility to improve the prediction accuracy of China's stock market volatility.From the perspective of high-frequency volatility in the international market,this paper first uses the principal component analysis method to extract the international volatility index for the volatility of the multinational(regional)stock market,and integrates the international stock market volatility information to forecast the volatility of China's stock market.Secondly,according to the mechanism transformation and time-varying characteristics of financial market fluctuations,on the basis of adding the international stock market high-frequency fluctuation information HAR-RV-X model,the Markov mechanism transformation is further introduced to analyze the characteristics of the impact of international stock markets on China's stock market.Further,in order to make more effective use of the information of international market volatility to predict China's stock market volatility.We consider the mutual transmission and time-varying integrity of international stock market volatility information.The high-dimensional TVP VAR method was first applied to the prediction and analysis of China's stock market price volatility.Considering the time-varying parameters of international stock market volatility,the weights of international stock market fluctuations are assigned dynamically to calculate the weight probabilities of different variable models,and the weighted average method is used to predict the volatility of China's stock market.Finally,based on the empirical studies in the previous three chapters,we consider volatility forecasting as the key determinant of portfolio optimization,and explore the economic value of different volatility forecasting models from the perspective of portfolio and risk management.Hopefully,this paper can provide some guidance for China's financial regulatory authorities to formulate and implement relevant financial policies,and also provide more practical and feasible basis for investors in correct judgment of stock market risks and choosing reasonable investment strategies.The empirical results mainly include:(1)Adopting the principal component analysis method is conducive to accurately obtaining common information of stock market high-frequency fluctuations from the interconnected global stock market fluctuations.The index has a positive impact and statistical significance on future Chinese stock market volatility.More importantly,in the prediction of China's stock market volatility,the HAR-RV extension model incorporating the international stock market volatility index performs better than the compared models,including HAR-RV,Kitchen sink model and five combined methods.(2)By analyzing the characteristics of the impact of international stock market volatility information on China's stock market volatility under different volatility states,we found the Markov mechanism transformation model can capture and process global international stock market volatility information more effectively and improve forecasting capabilities.The statistical test shows that the transition probability matrix is significant,and the high fluctuation state shows a significantly higher fluctuation level than the low fluctuation state.Compared with the linear and time-varying parameter HAR-RV model,the MSHAR-RV-X model can significantly improve the point prediction accuracy and direction prediction accuracy of the Chinese stock market.(3)The high-dimensional TVP VAR method can well capture the potential spillover effects and time-varying characteristics between international stock markets.It is a new and more accurate method to characterize and predict China's stock market volatility through international stock markets.(4)By maximizing investor utility,the performance of the investment portfolio is assessed.The empirical results show that,firstly,the HAR-RV extended model based on principal component analysis can significantly improve the accuracy of variance prediction and produce a higher economy value than the comparative models(including HAR-RV,Kitchen sink model and five combined methods).Secondly,compared with the HAR-RV model with linear and time-varying parameters,the MSHAR-RV-X model can significantly improve the direction prediction accuracy of China's stock market,and the strategy formed on this basis can gain larger combined benefit.Last but not least,the new high-dimensional TVP VAR approach achieves the best results in both direction prediction and economic value in the application of portfolio risk management.