Research On The Correlation Of Stock Index Returns Based On Time Varying Copula Method

The financial assets correlation is an important issue in risk management,which can be effectively solved by the Copula theory.Investors attention is aroused as Shanghai and Hong Kong Stock Connect program's opening in 2014,at the same time,they keep a wait state considering the investment risk.How do describe the correlation between Shanghai and Hong Kong's stock assets? The problem will be solved by Copula function in the paper,the main work in these may be summarized as follows:(1)In the paper,real stock returns data in Shanghai composite index and the Hang Seng H shares AH index is selected,Copula functions of Archimedes and Elliptic are used to study their correlation.The method of semi-parameters is used to estimation Copula's parameter,the optional Copula function is determined by using K-S and Euclidean distance ways.Empirical results show that the Student t-Copula can well describe the correlation between stocks,which means Shanghai and Hong Kong stock are symmetrical tail correlation(2)Considering the financial market is dynamic development,time varying condition Copula is used to studied correlation between the above two indexes.Taking into account the complexity of determining the edge distribution of data,empirical distribution function is selected as the marginal distribution.The AIC and the BIC criterion are used to select the best Copula function.The empirical results show that time varying Copula can better describe the correlation than static Copula,RG Copula better fits to describe the data,and these two stock show upper tail correlation.