Dependence properties and Bayesian inference for asymmetric multivariate copulas

摘要: We study a broad class of asymmetric copulas introduced by Liebscher (2008) as combination multiple – usually symmetric copulas. The main thrust the paper is to provide new theoretical properties including exact tail dependence expressions and stability properties. A subclass obtained combining comonotonic studied in more detail.We establish further for this show that they are characterized an arbitrary number singular components. Furthermore, we introduce novel iterative representation general which de facto insures uniform margins, thus relaxing constraint Liebscher’s original construction. Besides, construction proves useful inference developing Approximate Bayesian computation sampling scheme. This inferential procedure demonstrated on simulated data compared likelihood-based approach setting where latter available.