Research On The Capital Asset Pricing Based On Markov Switching Model Of Gibbs Sampling

The capital asset pricing problem is a hot spot in the academia in recent years.It is not only involves the traditional investment,finance,insurance,etc,but also has a wide application in other industries.The traditional research method is primarily on the assumption that expected return under the premise of obeying normal distribution.But in the process of actual operation of financial time series usually present rush thick tail or not normality.In this paper,we use the method based on Gibbs Sampling structure embedded Markov Switching Model.Use of its advantages of dynamic structural problems in processing,to quantitative study the financial time series is non-stationary,effectively to deal with the peak of financial time series is thick tail or not normality.Then construct using Gibbs Sampling method is used to solve the four zones of the Markov Switching Model,to analyze the Shanghai stock 50 index as an example,this model can better describe the volatility of financial yield traits.The empirical part using the capital asset pricing based on Gibbs-Markov Switching Model to study the Shanghai 300 only five stocks and the VaR model is given,it is concluded that for the coefficient of the drivers of inconsistency and transition time conclusion.Investment choices for investors and market regulation provides effective scientific basis.