25/08/2017 - A survival model with Birnbaum{Saunders frailty for uncensored and censored cancer data - Palestrante: Vera Tomazella (UFSCar)

Seminário Conjunto UFSCar/ICMC

Data: 25/08/2017 - 14h00

Local: Sala de seminários do DEs-UFSCar

Palestrante: Vera Tomazella - DEs-UFSCar

Titulo: A survival model with Birnbaum{Saunders frailty for uncensored
and censored cancer data

Resumo: Survival models with frailty are used when additional data are non-available to explain the occurrence time of an event of interest. This non-availability may be considered as a random eect related to unobserved explanatory variables, or that cannot be measured, often attributed to environmental or genetic factors. We propose a survival model with frailty based on the Birnbaum{Saunders distribution. This distribution has been widely applied to lifetime data. The random eect is the \frailty", which is assumed to follow the Birnbaum-Saunders distribution and introduced on the baseline hazard rate to control the unobservable heterogeneity of the patients. We use the maximum likelihood method to estimate the model parameters. Examples with uncensored and censored real-world data sets illustrate the potential applications of the proposed model.

ATENÇÃO: no final do seminário os Coordenadores do PIPGES irão fazer uma reunião com
todos os alunos do programa.

Sua presença é muito importante. Compareça!