B1598
Title: Inference for copulas with two-piece margins
Authors: Jonas Baillien - KU Leuven (Belgium)
Irene Gijbels - KU Leuven (Belgium)
Anneleen Verhasselt - Hasselt University (Belgium) [presenting]
Abstract: Copulas provide a versatile tool in the modelling of multivariate distributions. With increased awareness of possible asymmetry in data, skewed copulas combined with classical margins have been employed to model these data appropriately. The reverse, skewed margins with a (classical) copula has also been considered, but mainly with skew-symmetrical margins. We focus on different types of skewed margins, namely the two-piece distributions. More specifically, we use the recently proposed quantile-based asymmetric family of distributions in given copula structures. For this combination, we provide statistical inference results in consistency and asymptotic normality for the Inference Functions for Margins estimator. A simulation study complements the theoretical results, and the practical usefulness is shown through some real data examples.
