Advances and Applications in Statistics
Volume 28, Issue 2, Pages 123 - 131
(June 2012)
|
|
BAYESIAN INDEPENDENT TESTING IN A BIVARIATE WEIBULL MODEL
Jangsik Cho and Yong S. Joe
|
Abstract: In this paper, we consider a two-components system in which the lifetimes have a bivariate Weibull distribution. We propose a Bayesian testing procedure for independence based on Bayes factor. We use a noninformative prior such as an improper prior for the parameters so that the prior is defined only up to arbitrary constant which affects the values of Bayes factors. Also, we compute the fractional Bayes factor (FBF) proposed by O’Hagan [13] to compensate for that arbitrariness. We compute the FBFs and select the highest posterior probabilities for the hypotheses, respectively. Additionally, we give a numerical example to illustrate our procedure. According to the results, FBF methodology for Bayesian independent testing is reasonable. |
Keywords and phrases: Bayesian testing, bivariate Weibull model, fractional Bayes factor, improper prior, posterior probability. |
|
Number of Downloads: 358 | Number of Views: 1167 |
|