Copula function selection criterion based on nonparametric kernel density estimation
Zhang Shi-ying · Journal of systems engineering · 2010
It is very important to depict the dependence among the financial assets accurately in modern financial risk analysis.So,one key issue is the choice of an appropriate Copula in a given set of Copulas to model the dependence among the financial assets.In this sense,kernel density selection criterion,a new Copula selection method based on nonparametric kernel density estimation is proposed.Furthermore,with the Monte Carlo simulation,the kernel density selection criterion is compared systematically with AIC and the selection criterion based empirical Copula.The main conclusions are that the Copula selection result with the new method is best and the method needs not to compute the density function of each Copula in the given set of copulas.