Bayesian inferences for the Birnbaum-Saunders Special-Case distribution

• Autor(a) principal: Nakamura, Luiz Ricardo
• Data de Publicação: 2016
• Outros Autores: Leandro, Roseli Aparecida, Villegas, Cristian
• Tipo de documento: Artigo
• Idioma: eng
• Título da fonte: Repositório Institucional da UFLA
• Texto Completo: http://repositorio.ufla.br/jspui/handle/1/13967
• topic: Generalized Birnbaum-Saunders distributions; Markov chain Monte Carlo; Metropolis-Hastings algorithm; Random number generator; Distribuições Birnbaum-Saunders generalizadas; Monte Carlo via cadeias de Markov; Algoritmo Metropolis-Hastings; Gerador de números aleatórios

In this paper, we discuss the estimation of the Birnbaum-Saunders Special-Case (BS-SC) distribution through the Bayesian approach considering its parameters independents, assuming gamma priors for both of them. As the full posterior conditionals do not have closed forms we use the Metropolis-Hastings algorithm to generate samples from the joint posterior distribution. We present a simulation study proposing the Markov chain Monte Carlo (MCMC) method as a random number generator, considering the cases where the BS-SC distribution has symmetric and asymmetric shapes. An application related to ozone concentration is presented in this paper using the described methodology.