Research And Empirical Analysis Based On The Volatility Of SSE 50 ETF

Volatility plays an integral role in asset and derivatives pricing,optimal portfolio selection,and the risks associated with holding various financial instruments.Therefore,scholars are constantly striving to find suitable models to characterize and predict volatility.With the rapid development of financial technology,the acquisition and preservation of high-frequency data has become more convenient.The traditional GARCH model and SV model cannot meet the needs of high frequency data research.In this context,the realized volatility has been favored by scholars because of its simple calculation and no need for parameter estimation.Based on the heterogeneous market hypothesis,this paper constructs a heterogeneous autoregressive realized volatility(HAR-RV)model for volatility research.Taking into account the impact of non-trading time information on volatility,the overnight yield is introduced to improve the original model to obtain the HAR-ARV model.Since the residual of the model is usually heteroscedastic,we introduce the GARCH model and the EGARCH model,and extend the model to the HAR-ARV-GARCH model and the HAR-ARV-EGARCH model.This paper selects the high frequency data of SSE 50 ETF every 5 minutes from March 24,2016 to March 29,2019 as the research object.Based on the adjusted realized volatility sequence,the HAR-ARV model,the HAR-ARV-GARCH model and the HAR-ARV-EGARCH model are constructed.The parameters of the three models are estimated,and then the loss function is used to compare and analyze the prediction results.The results show that the HAR-ARV model,HAR-ARV-GARCH model and HAR-ARV-EGARCH model have good prediction results.After comparison,the prediction effect of HAR-ARV-GARCH model is stronger than that of the other two models,which can better measure and predict the volatility of SSE 50ETF.Finally,we carry out further analysis based on the HAR-ARV-GARCH model,which proves the heterogeneity of investors in China's fund market.