Hybrid models with complex censoring and random effect

Detalhes bibliográficos
Ano de defesa: 2024
Autor(a) principal: Montoya, Javier Ramirez
Orientador(a): Não Informado pela instituição
Banca de defesa: Não Informado pela instituição
Tipo de documento: Tese
Tipo de acesso: Acesso aberto
Idioma: eng
Instituição de defesa: Biblioteca Digitais de Teses e Dissertações da USP
Programa de Pós-Graduação: Não Informado pela instituição
Departamento: Não Informado pela instituição
País: Não Informado pela instituição
Palavras-chave em Português:
Link de acesso: https://www.teses.usp.br/teses/disponiveis/45/45133/tde-29012025-162218/
Resumo: In recent decades in studies on disease progression, there has been a growing interest to address joint modeling, in which multi-state models are widely used, because they are a useful tool for understanding the complex paths that individuals can follow. In this sense, under the classical parametric approach, the same probability distribution is assumed for all transitions. In this thesis, we propose a practical and flexible approach that uses different families for each transition time, improving the standard error estimates. This method allows us to quantify the association between transition times, modeling it by including a random effect to take into account the observed transition times of the same experimental unit. Under the philosophy of univariate frailty, we can obtain information on the heterogeneity of individuals. We therefore propose a generalization of the classic multi-state model procedure. We include in this thesis parametric and non-parametric approaches to the distribution of random effects, and we also introduce a parameter in the frailty factor to test the heterogeneity hypothesis without restricting of parametric space for the null hypothesis together with the complex censoring structure. We carried out simulation studies to evaluate the maximum likelihood estimates and applied our method to a real data set, demonstrating its practical application and potential impact in the field of disease progression modeling.