The Realized NGARCH Model And Its Application

In recent years,with the widespread application of electronic transactions in the financial field,as well as the rapid development of information technology,volatility in financial markets is becoming increasingly competitive.How to more accurately predict the risk of assets caught people's attention.Estimating the volatility of asset returns is one of the key problems in predicting the risk of earnings.The estimation accuracy of the volatility is closely related to the hypothesis of the model and the collection frequency of financial data.In general,the more accurate the volatility of the model and data of collection frequency the higher,the volatility estimates better accuracy.Therefore,it is the main research objective of this paper to establish a model of volatility with good statistical properties by using high frequency financial data.The main research contents and innovation points of this paper are as follows:1.The Realized NGARCH model is proposed by the NGARCH model reflects the leverage effect of volatility and introduced into parameters in equation of the Realized GARCH model.In the new model,the sequence of parameter with the errors becomes negative correlation.Innovations not only respond to the current returns in magnitude,but also respond to current earnings in direction.2.The parameter estimation directly affects the prediction accuracy of financial risk,so we use a Monte Carlo method to test the robustness of the parameter estimation of Realized NGARCH model.Stochastic simulation results show that the mean square errors of the all parameter estimation are significant in significance level.At the same time,the number of simulation is 500 times,the estimation of all parameters is still significant as the sample size increases.Simulation results show that,in this article the volatility estimation methods of the Realized NGARCH model have better robustness.3.Based on the new Realized NGARCH model,this paper makes an empirical analysis on the 5min high frequency financial data of the SSE 50 index and the SSE 380 index.The return risk predictor is tested by the ratio test.While,the prediction accuracy of the Realize NGARCH model and Realize GARCH have been compared.Empirical results show that the Realized GARCH model overestimates the risk of the market.In this paper,the Realized NGARCH model has been proposed to provide a new method to risk management,in a certain extent,which has enriched the theory of risk management in the financial field.