In recent years, Copula theory has grown extremely fast and extensive, particularly in the financial field. The copula theory provides a new point of view which we study the multivariate distribution. It reduces the choice of the best multivariate distributions into two steps:the margins selection, and the copula function selection.This article focuses on how to choose the Copula function, and the choosing copula methods are applied to the financial market dependence analysis and Value-at-risk calculation. The main work is as follow:First, we review the existed goodness of fit test methods and give two improved methods. Then we give four principles of Copula selection.Second, we analyze the dependence of Shanghai and Shenzhen stock market by using Copula selection method.Empirical result shows that there exists strong positive correlation of the two index return series.Third, the analytic formulae of dynamic VaR of portfolio are deduced based on the Copula-GARCH-GED model and the method of Monte Carlo simulation is proposed. Then, we use the data of Shanghai and Shenzhen stock market index return to do empirical analysis for dynamic VaR. Empirical analysis shows that the Copula choosing method and the Copula-GARCH-GED model are effective in reflecting the return of portfolio and estimating the dynamic VaR of portfolio.
|