Efficient goodness-of-fit tests in multi-dimensional vine copula models

摘要

We introduce a new goodness-of-fit test for regular vine (R-vine) copulamodels, a flexible class of multivariate copulas based on a pair-copulaconstruction (PCC). The test arises from the information matrix ratio. Thecorresponding test statistic is derived and its asymptotic normality is proven.The test's power is investigated and compared to 14 other goodness-of-fittests, adapted from the bivariate copula case, in a high dimensional setting.The extensive simulation study shows the excellent performance with respect tosize and power as well as the superiority of the information matrix ratio basedtest against most other goodness-of-fit tests. The best performing tests areapplied to a portfolio of stock indices and their related volatility indicesvalidating different R-vine specifications.