Title: Dependence properties and Bayesian inference for asymmetric multivariate copulas
Authors: Marta Crispino - University of Oslo (Norway) [presenting]
Stephane Girard - Inria (France)
Julyan Arbel - Inria (France)
Abstract: Some new theoretical properties of a broad class of asymmetric copulas will be introduced. Such copulas are obtained as a combination of multiple, usually symmetric, copulas. We will also focus on a subclass of Liebscher copulas obtained by combining comonotonic copulas which are characterized by an arbitrary number of singular components. Furthermore, we will introduce a novel iterative representation for general Liebscher copulas which de facto insures uniform margins, thus relaxing a constraint of the original construction. This iterative construction proves useful for inference by developing an Approximate Bayesian computation sampling scheme. The inferential procedure will be demonstrated on simulated data.