Copula-based bivariate finite mixture regression models with an application for insurance claim count data

Abstract Modeling bivariate (or multivariate) count data has received increased interest in recent years. The aim is to model the number of different but correlated counts taking into account covariate information. Bivariate Poisson regression models based on shock approach are widely used because their simple form and interpretation. However, these do not allow for overdispersion or negative correlation, thus, other have been proposed literature avoid limitations. present paper proposes copula-based finite mixture models. These offer some advantages since they all benefits a mixture, allowing unobserved heterogeneity clustering effects, while derivation can produce more flexible structures, including correlations regressors. In this paper, new defined, estimation through an EM algorithm presented, then applied Spanish insurance claim database.