Title: Copula additive regression models with endogenous binary treatment and count response
Authors: Giampiero Marra - University College London (United Kingdom) [presenting]
Rosalba Radice - Cass Business School (United Kingdom)
David Zimmer - Western Kentucky University (United States)
Abstract: Copula regression models are discussed for a count response and an endogenous binary treatment, where the marginals and copula function can be chosen from a rich set of distributions and all the models parameters can be flexibly specified as functions of additive predictors. Estimation is achieved using a simultaneous penalised likelihood approach with automatic multiple smoothing parameter selection. Inferential results are also briefly discussed. The modelling framework is implemented in the R package GJRM (Generalised Joint Regression Models). The approach is illustrated on a case study which investigates the effect of insurance status (a binary measure) on doctor visits (a count measure).