Title: A simple non-parametric goodness-of-fit test for elliptical copulas
Authors: Miriam Jaser - Technical University of Munich (Germany) [presenting]
Aleksey Min - Technische Universitaet Muenchen (Germany)
Abstract: A simple non-parametric goodness-of-fit test for elliptical copulas of any dimension is presented. It is based on the equality of Kendall's tau and Blomqvist's beta for all bivariate margins of an elliptical copula. First, the asymptotic distribution of the test statistic is derived under the assumption of known marginal distributions. Nominal level and power of the proposed test are investigated in a Monte Carlo study. In case of unknown marginal distributions, we estimate the margins non-parametrically and derive the limiting distribution using empirical copula processes. The limiting Gaussian field depends on the unknown copula. Therefore, we apply bootstrap procedures to conduct the goodness-of-fit test. Finally, an empirical application illustrates our goodness-of-fit test at work.