Local Polynomial Estimation Of Copula Function And Its Correlation In The Stock Market

Emerging financial crisis in contemporary society, frequent fluctuations in financial markets intensified, the correlation between markets is also more complex, including the relevance of these complex non-linear correlation, correlation and non-symmetrical tail correlation, the correlation structure of the different basic and comprehensive description of the relationship, and different forms of composition related to different correlation structure, each with characteristics related to structure. Currently Copula function has been widely used in financial markets, because it avoids some of the drawbacks of the traditional correlation particularly in the financial markets risk measurement, portfolio, asset pricing, etc., so it has become the financial markets correlation an effective tool. Some of the traditional measure of the relevance of more or less have their own limitations and drawbacks of using Copula function to model the relevance of finance can effectively overcome this, it is to better characterize the nonlinear variables, Asymmetry and tail dependence and other characteristics, can be more extensive analysis of the correlation is also more practical. Copulas function with the emergence of the traditional one or two indicators to represent the correlation structure of the method with a more complete and comprehensive representation of the correlation between variables, the hair replacement methods. Copula function can therefore not decide whether the linear correlation coefficient of the correct measure of relevance of the premise down. The Copula function of the parameter estimation method is to model the Copula function is a crucial research question. Copula functions commonly used estimation methods are maximum likelihood estimation, moment estimation, nonparametric methods, but these are more or less common estimation method has some limitations. Local polynomial method can estimate a wide range of financial assets to use the density distribution, local polynomial estimation method uses the assumption of assets find more dependent structure, while addressing the asset structure of the density distribution of the dependence problem is One way the United States both worlds, and can accurately analyze the distribution of financial assets related to the structure of the two. Effectively avoid the maximum likelihood estimation of the limitations of such reasoning.In this paper the theory of nonlinear programming is introduced into the local polynomial estimation Copula functions and through empirical proof of its availability and effectiveness.