L. Sanz San Miguel, M. A. Gómez Villegas, J. M. Marín Diazaraque
Reference analysis is used to produce Bayesian inferences based only on the assumed model and the available data. In this context, the prior distribution used to make inferences about a parameter of interest is dominated by the data and is less informative regarding some information criteria. Here we obtain reference prior distributions for parameters of the univariate family of power exponential distributions. We also obtain reference posteriors using reference priors and the Bayes theorem. The results are checked via simulation examples and a real data example and are applied in carrying out a test of normality.
Palabras clave: Reference prior; Fisher information; exponential power distribution; hypothesis testing; likelihood; repeated measurements
Programado
L08.2 Grupo de Inferencia Bayesiana: Applications of Bayesian Modelling in Econometrics and Genetics
5 de septiembre de 2016 15:40
0.09 - Aula de proyectos 2