Use of bayesian methods in accelerated life tests assuming an exponentiated-Weibull distribution
and a power inverse law model

Denilton da Silva VIEIRA1

Jorge Alberto ACHCAR[1]

Vicente Garibay CANCHO[2]

§    ABSTRACT: In this paper, we explored the use of Chains Markov Monte Carlo (MCMC) techniques to develop a Bayesian analysis for accelerated life tests considering the Weibull-exponentiated distribution. Assuming type II censored data, we considered a classic analysis and a Bayesian analysis assuming informative priors for the parameters of the model, where we obtained the predictive density for a future observation and proposed a criterion that can be used in quality control problems. We illustrated the  methodology with a numeric example.

§    KEYWORDS: Exponentiated-Weibull distribution; accelerated life tests; predictive densities; MCMC.



[1]Departamento de Estatística, Universidade Federal de São Carlos - UFSCar, CEP 3565-905, São Carlos, SP, Brasil. E-mail: denisnog@yahoo.com.br / achcar@power.ufscar.com.br

[2]Departamento de Matemática Aplicada e Estatística, Instituo de Ciências Matemáticas e de Computação, Universidade de São Paulo - USP, Caixa Postal 668, CEP 13560-970, São Carlos, SP, Brasil. E-mail: gariby@icmc.usp.br