Estimação Bayesiana dos parâmetros da distribuição Exponencial Generalizada Bivariada

• Autor(a) principal: Coladello, Leandro Fernandes [UNESP]
• Data de Publicação: 2014
• Tipo de documento: Dissertação
• Idioma: por
• Título da fonte: Repositório Institucional da UNESP
• Texto Completo: http://hdl.handle.net/11449/108637
• topic: Computação - Matematica; Funções hiperbolicas; Teoria bayesiana de decisão estatistica; Computer science - Mathematics

Many bivariate distributions for survival analysis were proposed, but the Bivariate Generalized Exponential Distribution (BVGE) presented by Gupta and Kundu (2009) has interesting properties. For example, the BVGE distribution has Generalized Exponential marginal distributions, which is used in many unidimensional problems. An alternative way to obtain multivariate (or bivariate) distributions is the use of Copula theory, which is proving to be a useful alternative, because it permits the construction of multivariate distributions without the necessity of giving restrictions to marginal and multivariate distributions. Inference about different bivariate models of failure time are very important, and consequently, comparisons can be made and the Bayesian method is recognized to offer significantly benefits in data analysis, justifying its use. This work considered comparisons between the Bivariate Generalized Exponential Distributition and generalized bivariate exponential distributions obtained by copulas functions.