Studies On The Stability And Time Variation Of Industrial Betas In Shenzhen Stock Market

The systemic risk in traditional capital asset pricing model(CAPM) is completely refected through beta coefficient.The traditional CAPM is a single-period equilibrium model,the systemic risk described by it is static,it does not take into account the fact that investors' strategies are changing when they make investment decisions.Therefore,it has limitation to measure the dynamic equilibrium of the capital asset market.This thesis uses the industry indixes and the composite index of Shenzhen Stock Market to compute the returns of every industry and the market.Based on the studies of the stability and time variation of betas,we try to solve the following main problems:(1)whether the systemic risk of every industry can be measured by static betas;(2)if the betas are unstable,whether the static CAPM can be converted to dynamic CAPM;(3)if the answer in(2)is yes,whether the dynamic systemic risk can be fitted and forecasted well.In the thesis,the main employed methods to identify the stability of betas are based on the residuals test of OLS,the test of betas of rolling regression and the CUSMSQ statistics of recursive regression;the discriminate method of the time variation of betas is Kalman filter analysis based on state-space models,and the choosen models are random walk model,random coefficient model and mean-reverting model.The research results show that:(1)the betas of all industries in Shenzhen stock market are unstable;(2)the betas of all industrie in Shenzhen stock market have time-varity;(3)in the three used models,the betas often out of thirteen industries in Shenzhen Stock Market is fitted best by mean-reverting model,and the betas of the other three industries is fitted best by random coefficient model,however,the random walk model is not suitable to fit the betas of any industry.The means value of betas in all industries are not equal to 1 and not same,the systemic risk in most industries are bigger than the market risk;(4)compared to static CAPM,the dynamic CAPM has smaller errors in forecasting.The thesis studies the stability and time variation of the betas in all industries of Shenzhen Stock Market from the perspective of empirical analysis,the conclusions show the limitation of traditional CAPM and provide a good reference for the further studies of the theories and applications of CAPM.