The Risk Correction Of F.F Five Factor Model System Based On Two-stage Copula Bayes

At present,with the development of global financial system integration,more frequent financial crises and faster transmission speed are followed.Therefore,it is necessary to measure systemic risk effectively.However,due to the impact of asset structure and market friction,as well as other factors such as investment term,the systemic wind is caused.The measurement of risk has become a difficult problem that many scholars in the financial field want to optimize.The CAPM model developed by Sharp et al.in 1964 is a model for early scholars to study the return and risk in the securities market.After a series of development and evolution of CAPM model,Fama and French expanded it into a three-factor model in 1992.In 2015,Fame and French added profit factor RMW and investment factor CMA to the three-factor model to form a five-factor model which is more in line with the stock market trend.Five-factor model has attracted extensive attention and argumentation.Scholars at home and abroad have used five-factor model to verify its fitting relationship with the stock market.At present,a large number of academic achievements show that the five-factor model is superior to the three-factor model in explaining the stock market returns.Therefore,this paper adopts F.F five-factor model.The prime model explores the estimation of systemic risk beta,relaxes the assumption of fixed investment term,adds the investment term as an uncertain parameter to the model,and then uses two Bayesian copula estimates to observe the relationship between systemic risk beta and investment scale,and adds performance between Chinese and American industries.By contrast.This paper selects the monthly data of Chinese and American stock markets as the data base.The data of Chinese stock market is divided into two periods:from January 2001 to December 2008 and from January 2009 to December 2017.The data of American stock market is from January 2009 to December 2017,and the industries are divided according to CICS.According to the industry classification,the data of non-daily,industrial,public utilities,daily necessities,information technology and medical services were selected from China and the United States respectively.The investment scale factor was added into the F.F five-factor model,and the SUR regression method was used to regression the sub-industries of China and the United States respectively.The copula was used twice again.A Bayesian estimation is used to get investment scale and systemic risk.Through empirical analysis,the first analysis from the regression results,the overall factor coefficients of China and the United States have a certain significance.The market value scale and book-to-market ratio in China's stock market have significant effects,and also have certain profitability effects and investment style effects.The profitability and investment style effects of the US market are significantly higher than those of the Chinese market,and the FF five-factor model is more significant in the US market.Second,from the perspective of the relationship between the systemic risk of Sino US industries and the sensitivity of investment targets.There is no obvious correlation between the two markets.With the change of investment scale,the beta value of systemic risk has no obvious positive correlation or negative correlation.