AN APPROXIMATE BAYESIAN ANALYSIS FOR ACCELERATEDTESTS WITH LOG-NON-LINEAR STRESS-RESPONSE RELATIONSHIP

Gleici Castro PERDONÁ[1]

Francisco LOUZADA-NETO[2]

§       abstract: In this paper we propose an accelerated life tests model with a log-non-linear stress-response relationship. The advantage of  such a formulation is that the general framework accommodates several stress-response relationships usually considered in accelerated life tests, while it is flexible enough for fitting the data that cannot be accommodatet by a simple log-linear stress-response relationship. A Bayesian approach with constant prior density is considered for interval estimation by using Laplace's method to find the marginal posterior densities of interest. Model comparation is made through Bayes factors. The methodology is illustrated in a numerical example.

§       keywords: Accelerated lifetime tests; Bayesian analysis; exponential distribution; Laplace method; log-non-linear stress-response relationship..



[1] Departamento de Medicina Social, Faculdade de Medicina de Ribeirão Preto - FMRP, Universidade de São Paulo - USP, CEP 14049-900, Ribeirão Preto, SP, Brasil. E-mail: pgleici@fmrp.usp.br

[2] Departamento de Estatística, Universidade Federal de São Carlos - UFSCar, Caixa Postal 676, CEP 13565-905, São Carlos, SP, Brasil. E-mail: dfln@power.ufscar.br