Volatility Spillover Effect Between Stock Markets Based On Copula Model With Structural Changing In Tail

With the rapid development and opening up of Chinese economy, the economy of China gets more and more close with the world economy. The Chinese stock market is deeply affected by the risk contagion effects of world main stock market. In recent years, with the frequent financial volatility and financial crisis, the volatility spillover effect between the financial markets get more and more close attention by scholars at home and abroad. The traditional correlation measurement method which only measures the linear correlation between variables has been increasingly unable to meet the needs of people to describe the correlation.Under this case, this paper introduces the measurement tools that can measure nonlinear relationship—the Copula function, which can establish the nonlinear correlation model.Previous researches mostly use a single Copula function, which can not fully describe the dependence structure of financial variables. Hybrid Copula function which contains different types of Copula can measures more flexibly dependency between variables. Thus, this paper constructs a mixed copula model which possesses with the changing tail structure by linear combination method. The hybrid copula can simultaneously measure the variables between the upper and lower tail dependence and also can combine the measurement of correlation degree and correlation model. In this paper, another innovation is that we consider the time-varying weight parameters of the mixed copula model which capture the dynamic dependent model between variables,while the copula function parameters are constant.In this paper, we choose Shanghai Composite Index, Hong Kong Hang Seng Index, America S&P500index and the Nikkei225index from2000January to2013January to study dependency changes between Chinese stock market and the world’s major stock markets and to estimate the existence of risk spillover effect. Firstly we use AR(1)-GARCH(1,1)-t model to simulate the marginal distribution index. Secondly we use the static Normal Copula, Gumbel Copula, Clayton Copula, SJC Copula function and the time-varying hybrid Copula function we construct to model risk contagion between the Chinese stock market and international stock markets and to analysis the empirical results. The empirical results show that in the process of opening up of Chinese capital market, linkage between Shanghai index and main international indexes is not strong. The Shanghai Composite Index has weak influence on main international stock markets. Although China’s stock market has low tail dependence with the four other stock markets. The markets still have the possibility of same change trend when facing the extreme events, That,to some extent,reflect the global stock market’s fluctuation by big financial crisis,especially caused by the United States of America.