Research On Measurement And Prediction Of Industry Tail Risk Based On Bayesian Expectile Regression

Financial risk management plays an important role in modern economy,financial market and investment decision-making.In financial risk management,it is crucial to find scientific and accurate risk measurement methods.In the stock market,industry risk is an important research direction,which is of great significance to the financial market and real economy investment.Thus,this paper proposes research on the measurement and prediction of industry tail risk.In view of the fact that traditional quantile regression method cannot fully describe the tail characteristics of different financial data distributions,this paper proposes a Bayesian expectile regression model.The model uses expectile,which is more sensitive to the thick tail characteristics of the distribution,to measure and predict tail risk.In terms of parameter estimation of the model,Bayesian ideas are introduced and the sampling steps of the M-H sampling algorithm embedded in the Gibbs sampling framework are designed based on the MCMC method to estimate uncertain parameters.This paper also uses the Bayesian expectile regression model to conduct empirical research on the closing price data of the CSI 300 index and CSI 300 primary industry indexes and draws conclusions on the robustness of the model,measurement effectiveness,prediction performance and industry differences.The results show that: firstly,the Bayesian expectile regression model has significant advantages over the Bayesian quantile regression model in tail risk measurement and prediction,and the model itself has good robustness and adaptability.Secondly,the measurement advantages of the model vary across different industries,with the best measurement performance in the pharmaceutical industry.Finally,market fluctuations represented by historical income information have a consistent impact on various industries.Specifically,all industries are more sensitive to market positive information,but there are obvious industrial differences in this impact.Based on this,11 primary industries can be divided into three categories,with similar industries being similarly affected by market shocks.