Advances and Applications in Statistics
Volume 17, Issue 1, Pages 1 - 27
(July 2010)
|
|
USE OF MCMC METHODS TO OBTAIN BAYESIAN INFERENCES FOR THE BIRNBAUM-SAUNDERS DISTRIBUTION IN THE PRESENCE OF CENSORED
DATA AND COVARIATES
Jorge Alberto Achcar and Fernando Antônio Moala
|
Abstract: The use of the Birnbaum-Saunders distribution for a lifetime T or the sinh-normal distribution for the logarithm of T could be a good alternative to analyse lifetime data. We introduce a Bayesian analysis using MCMC (Markov Chain Monte Carlo) methods to simulate samples for the joint posterior of interest assuming the Birnbaum-Saunders distribution or the sinh-normal distribution. A great simplification is obtained using the WinBugs software. We also consider linear regression models assuming a sinh-normal distribution for the error in place of the usual normal distribution. Some examples are introduced to illustrate the proposed methodology. |
Keywords and phrases: Birnbaum-Saunders distribution, sinh-normal distribution, MCMC methods, linear regression models. |
|
Number of Downloads: 360 | Number of Views: 879 |
|