The Construction And Empirical Analysis Of Rational GARCH-Exp Model Based On Quantile Regression

With the development of China's financial market,the stock market is playing an increasingly important role in the financial market.Volatility is an important index to measure financial risks.It is estimated that volatility can predict future trends,provide investors and managers with more accurate choices of prevent risks.In addition,the stock price index can measure and reflect the overall price level and changing trend of the stock market within a certain range,which is of great reference significance to investors' market price judgment.Based on this,this paper takes SSE 380 index and shenzhen component index as the research data and conducts empirical analysis.most financial assets exhibit asymmetric volatility,so this paper uses the Rational GARCH-Exp model to describe asymmetric volatility.Secondly,the peak and thick tail characteristics based on financial data are compared with the standard normal distribution,so this paper will evaluate the performance of the Rational GARCH-Exp model in estimating the return risk by comparing the normal distribution,the student-t distribution,and the generalized error distribution.Finally,this paper calculated the VaR value of QR-Rational GARCH-Exp model based on quantile regression method and compared it with the VaR value of Rational GARCH-Exp model based on parameter estimation method,and conducted the failure rate test.The empirical results show that the asymmetric Rational GARCH-Exp model has a better prediction effect on volatility than the classical GARCH(1,1)model.Among them,the Rational GARCH-Exp model based on generalized error distribution(GED)is more effective in predicting volatility than the model based on normal distribution and student-t distribution.In this paper,the VaR value obtained by QR-Rational GARCH-Exp model established by quantile regression method is better than that obtained by Rational GARCH-Exp model.