A bivariate Weibull regression model
for survival data

Gladys Dorotea Cacsire BARRIGA[1]

Francisco LOUZADA NETO1

Vicente Garibay CANCHO[2]

§    ABSTRACT: In this paper, we propose a new  bivariate Weibull model  (BVW), that it allows to model dependent survival data. The proposed model is motivated by an application of a result of Morgenstern (1956). An extension of this new model is also considered adding covariates to the model BVW. A Bayesian methodology  based on methods of Monte Carlo  Chains of Markov is developed for the proposed models. We illustrated the methodology with three set of real data.

§    KEYWORDS: Distribution bivariate; bivariate failure time; bayesian inference; bivariate Weibull distribution.

 



[1] Departamento de Estatística, Universidade Federal de São Carlos – UFSCar, CEP:13565-905, São Carlos, SP, Brasil.. E-mail: dfln@power.ufscar.com.br

[2] Departamento de Matemática Aplicada e Estatística-ICMC, Universidade de São Paulo – USP, CEP:13560-970, São Carlos, SP, Brasil. E-mail: garibay@icmc.usp.br