The Research Of Har-RV Model On The Forecasting Ability Of Stock Market Rolatility In China

Volatility refers to the degree of volatility of the price of financial assets.It can be used to measure the uncertainty of the return on assets,reflect the risk level of financial assets,and provide a reference for market participants and decision makers to analyze and predict the future market trend.On November 28,2016,the sse 50 ETF volatility index(abbreviated as China wave index or IVIX index)was officially released by China securities index co.,LTD.,which is also known as "China's version of the fear index".It can be seen that it is of great significance to identify and manage market risks by selecting a suitable prediction model to predict the volatility of China's stock market.In view of the previous analysis and research,the stock volatility of emerging countries has the characteristics of long memory,while the stock market of China has strong heterogeneity.On this basis,this article summarized the domestic and foreign literature about the stock market volatility characteristics,on the basis of research,model research,selecting Shanghai50 stocks index as sample data,by getting the minutes level index data,select 5 minutes for calculating the optimal sampling frequency high frequency earnings for realized volatility,the heterogeneity of regression(HAR)model on Chinese stock market volatility forecast ability.On this basis,on the basis of the original model,this paper also adds the financial market liquidity index turnover to constitute the HAR-RV(V)model.The ability of the HAR-RV model and the original model to predict daily,weekly and monthly volatility of the Shanghai stock exchange 50 market was compared.This paper analyzes whether the volume as an important flow index can improve the ability of the model to estimate and predict volatility.According to the empirical results of this paper,the following conclusions are drawn:first,the realized volatility is more accurate than the daily volatility rate of return,and has the characteristics of log-normality and long memory.Second,HAR-RV has a strong ability to predict the short-,medium-and long-term volatility of sse 50 index.Thirdly,the HAR-RV(V)model composed of a liquidity metric(volume)can improve the prediction ability of the overall volatility to some extent.