Volatility Study And VaR Evaluation Of SSE 50 ETF

In modern financial markets,volatility plays an important role in financial products pricing,trading and risk management.In this paper,the realized GARCH model is established by using the 5-minute high-frequency data of SSE 50 ETF,and the volatility of the ETF and the VaR are analyzed empirically.In terms of volatility estimation,the realized GARCH model is compared with the traditional GARCH model,and the realized GARCH models abased on normal distribution,t distribution and Skewed-t distribution under three different distributions are compared.In terms of VaR,the one-step prediction method based on rolling window is used to predict the VaR.The Kupice failure test is used to compare the prediction effect of different models.The empirical results show that the series of the return of SSE 50 ETF has the features of leptokurtosis and fat tail,skewed,and volatility clustering;the realized GARCH model combining the realized measure and the conditional volatility is superior to the traditional GARCH mode;Realized GARCH model with the skewed student's t distribution performs better than that with the normal and standard t distributions.;Realized GARCH model with Skewed-t distribution which has skew and thick tail features can better predict the risk value.