"A goodness-of-fit test for regular vine copula models" und "Efficient goodness-of-fit tests in multi-dimensional vine copula models"

Ulf Schepsmeier (TU München)
Monday, April 14, 2014 - 2:00pm
Spandauer Straße 1, Room 23

In this talk we introduce several goodness-of-fit (GOF) tests for regular vine (R-vine) copula models, a flexible class of multivariate copulas based on a pair-copula construction (PCC). They are build hierarchically on (conditional) bivariate copulas only. In particular we investigate two new goodness-of-fit tests arising from the information matrix and specification test proposed by White (1982) and the information ratio test by Zhang et al. (2013). The test statistics are derived and their asymptotic distribution proven. Further 13 GOF tests are adapted from the bivariate case and compared in an extensive power study, which shows the superiority of the information matrix based tests. The bootstrapped simulation based tests show excellent performance with respect to size and power, while the asymptotic theory based tests are inaccurate in higher dimensions. The best performing GOF tests are applied to a portfolio of stock indices and their related volatility indices validating different R-vine specifications.