The Topic:Correlation Analysis Of Financial Market Volatility Based On Vine-copula

A new research method in the research field of Copula theory is the structure of Vine model based on Vine-copula decomposition in recent years. It is used in multivariate random variables in data modeling, using a series of Vine-copula module in the structure based on the graphic modeling tool Vine. Normally, use it to describe the multivariate joint distribution and analyze the dependence structure between multiple variables, provides the theoretical basis for the analysis case of high dimensional data. With the multivariate Copula function to describe the multivariate joint distribution compared, Vine-copula decomposition model is the multivariate joint distribution is decomposed into Vine-copula module series between two variables, together with Vine structure. Each of the Vine-copula module is two variables Copula function based on the AIC value to select the appropriate to be fit, which makes the modeling flexibility to highlight the difference between variables, multivariate distribution can be used to capture in this paper, the relevant basic knowledge Vine-copula were certainly introduced, including the definition of Copula function, the basic nature of the Copula function, several common Copula function, defined Vine, rules vine definition and so on.The first chapter in this paper briefly describes the research background and research significance of the relevant theoretical knowledge, to the domestic and foreign research status are presented recently, at the same time, describes the structure of the paper and the main contribution.In the second chapter, the basic knowledge of Vine-copula is introduced, including the definition of Copula, the Copula function, the basic properties of several commonly used Copula function, the definition of Vine, the definition of the regular vine.The third chapter introduces Vine-copula model to construct a complete for modeling step multivariate the relevant structure of financial market, and two for nonlinear parameter estimation algorithm (two times planning algorithm of gradient method and sequence), and by fitting the simulation performance of two types of algorithms to compare, to analyze the choice of algorithm will have an impact the validity of the fitting results.The fourth chapter in this paper USA subprime mortgage crisis this special period of time as the time point, combined with the Vine-copula model for four (Chinese financial market, stock market, Hongkong stock market, the Japanese stock market, the stock market USA) the empirical analysis between the influence of fluctuations, American subprime crisis on other financial markets whether it will cause a certain degree of.