Estimation And Applications Of VaR Based On Dynamic Copula

The analysis of the dependence is a key problem in multivariate financial analysis. It also plays an important role in measure portfolio risk.Traditional research methods of correlation analysis,which just use the linear correlation to describe the dependence structure between random variables is incomplete.Recently,Copula theory has been introduced to the financial field.Copula functions can combine the degree of dependence and dependence model,and it measures the value at risk precisely.It is noted that most studies assumed that the dependence is constant in the period,including the degree of dependence and dependence model.In fact,the financial relationship between the assets will change over time.Copula theory,with its Advantages of describing the relationship between non-linear modeling and flexible way to describe the dependencies between financial assets has been widely applied.In order to catch the dynamic changes of correlation for good measuring of portfolio risk,the key points and main content of this work are listed as follows: Firstly,a basic theory of the Copula theory was introduced and introduced the advantages of Copula function at describing the relationship between financial assets, and introduces its multi-time series dependences application.The paper discusses the researches on dynamics Copula theory recently,and introduces the procedure of modeling the dynamic Copula function and GARCH model.The paper introduces the basic theory and estimation methods about value at risk,then discusses the model VaR with the methods of Copula-GARCH.Finally,Empirical analysis was used for the Rate of Return for indices of Shanghai and Shenzhen Stock market.With Monte Carlo simulation techniques,paper estimates the value at risk of the portfolio.It was found that dynamic dependency modeling using time varying T-Copula can more accurately measure the market portfolio risk.