Research On Systemic Risk Forecast In China's Stock Market

In the stock market,how to measure risk and predict the future trend of stock more accurately has always been a hot issue in the economic and financial field.Stock risk can be divided into two kinds,systemic risk and non-systemic risk.Systematic risk can not be offset by portfolio investment compared with non-systematic risk.Therefore,if the prediction accuracy of future systemic risk in stock market is improved,investors can choose their own portfolio according to their own assets.This paper focuses on the prediction of systemic risk in stock market.First,Empirical Mode Decomposition(EMD)is used to decompose the closing price index series of stock market according to different fluctuations according to the basic principle of this method.Then,according to the characteristics of the decomposed series,appropriate prediction models are selected to predict.Finally,the prediction results are integrated and compared with the traditional generalized autoregressive conditional heteroscedasticity model(GARCH).This paper is mainly divided into two parts for empirical comparative analysis.In the first part,two representative stock index series,Shanghai Composite Index and Shenzhen Composite Index,are selected.EMD decomposition method and traditional GARCH prediction model are applied to predict the closing price of stock,and the relationship between the predicted value and the real value of the original series is compared.In the second part,the above two stock index series are selected to predict the systemic risk of the stock market,and VaR is selected as the index to measure the systemic risk.The relative error of VaR under the two methods is compared.The empirical comparison results show that in the first part,compared with the traditional GARCH model,the predicted value obtained by EMD decomposition method is closer to the real value of the closing price index,so it is concluded that the prediction accuracy can be greatly improved by using EMD decomposition method.In the second part,on the systemic risk of stock market,the value of VaR obtained by EMD decomposition method is closer to the value of VaR obtained by real sequence.Therefore,the application of EMD decomposition method can improve the prediction accuracy to a certain extent and obtain more accurate prediction results,whether in the prediction of closing price index or in the prediction of systemic risk.