On multivariate asymmetric dependence using multivariate skew-normal copula-based regression

In this paper we propose a procedure to study the asymmetric dependence of the multivariate data. The proposed procedure comprises methodologies that have not been considered in the analysis of multivariate asymmetric dependence. We first utilize the asymmetric multivariate copula-based regression to capture the asymmetric dependence among multiple variables. We then introduce the multiple asymmetric dependence measure to quantify the asymmetry in the predictive power of the tentative predictors for a tentative response variable. We demonstrate the proposed methods using a class of asymmetric multivariate skew normal copulas. An application example on the asymmetric comovements of financial assets illustrates the benefits of the proposed methods. (C) 2017 Elsevier Inc. All rights reserved.