Modeling of Correlated Data and Multivariate Survival Data

Abstract

The main goal of this thesis has been the development of models for different types of complex structures, all of which involving dependent outcomes. The first chapters were devoted to the study and modelling of continuous data within the framework of surrogate endpoints. We contributed to the modelling of correlated survival data, using a Plackett copula and a set of inferential tools. Towards the end of this thesis, we studied models with crossed random effects in different settings and we proposed an alternative method for the estimation of the effects in particular situations.