A new class of copula regression models for modelling multivariate heavy-tailed data

A new class of copulas, termed the MGL copula class, is introduced. The originates from extracting dependence function multivariate generalized log-Moyal-gamma distribution whose marginals follow univariate (GLMGA) as introduced in Li et al. (2021) . can capture nonelliptical, exchangeable, and asymmetric dependencies among marginal coordinates provides a simple formulation for regression applications. We discuss probabilistic characteristics obtain corresponding extreme-value copula, named MGL-EV copula. While survival be also regarded special case MGB2 Yang (2011) , we show that proposed model effective modelling structures. Next to simulation study, propose two applications illustrating usefulness model. This method implemented user-friendly R package: rMGLReg