Title: Copula bivariate generalized survival models
Authors: Rosalba Radice - Cass Business School (United Kingdom)
Giampiero Marra - University College London (United Kingdom) [presenting]
Abstract: Copula bivariate generalized survival models are described, where the parameters of the marginal survivor functions and copula dependence are modelled as functions of additive predictors. The marginals include proportional hazards and proportional odds models. Estimation is carried out using an efficient and stable algorithm that allows for automatic estimation of the smoothing parameters. The models are implemented in the R package SemiParBIVProbit and are illustrated on a medical case study.