Research On Systematic Risk Prediction Of Bank Stocks

The systemic risk of bank stocks refers to the influence of the whole political,economic and social environmental factors on the price of bank stocks,including policy risk,cyclical fluctuation risk,interest rate risk,purchasing power risk and exchange rate risk.The systemic risk of China's bank stocks is still relatively high,and it can't be dispersed through the investment portfolio.Therefore,It is very necessary to study the systemic risk of bank stocks.Through the review of existing literature,we find that there is no unified conclusion among scholars at home and abroad about the systemic risk's influence factors research of the bank stocks,and in the new historical period,the systemic risk of bank stocks may exhibit new characteristics;in the aspect of the systemic risk prediction,there are still some flaws in the forecasting model,and the traditional forecasting model is used to predict systemic risk with the relatively large error and poor forecasting accuracy.In view of these deficiencies,the paper uses beta coefficient to measure systemic risk,analyzes systemic risk's influence factors research of bank stocks.The paper introduces the network search data,and constructs systemic risk prediction model of bank stocks based on state space model.The main work and innovative achievements of this paper are as follows:In terms of research on systemic risk's influence factors of bank stocks,according to the characteristics of the new historical period,the paper selects 15 influence factors mainly from the macroeconomic environment,the characters of commercial banks and performance of the bank stocks in the stock market.Through factor analysis,the paper extracts4 main factors,and establish the multiple regression model.It is found that,among the 4 main factors,loan factor has a significant positive impact on beta coefficient extracted.Real estate investment factor and GDP factor have a significant negative impact on beta coefficient,and M2 factor has not passed the significance test.In terms of the systemic risk's forecasting research,the paper takes 16 banks' systemic riskas the research object.Based on research on systemic risk's influence factors of bank stocks,the paper introduces the network search data to fit and predict bank stocks' beta coefficient.The root mean square error RMSE and the mean absolute percentage error MAPE are compared with the traditional prediction model,the results show that the beta coefficient forecasting model based on the network search data is less error than the traditional prediction model,and the beta coefficient forecasting model based on the network search data is better fitted.This paper has great theoretical and practical value in the research of the influence factors and prediction of bank stocks' systemic risk.This paper analyzes the systemic risk's influence factors of the bank stocks,which explains the difference of the systemic risk of the bank stocks,and helps the regulatory authorities to prevent the systemic risk,maintain the market order and ensure the healthy and stable development of the market.The systemic risk prediction model constructed by this paper can provide investors with decision-making support,guide investors to invest correctly and rationally,avoid investment risks and reduce investment losses.