3rd International Symposium on
Imprecise Probabilities and Their Applications

ISIPTA '03

University of Lugano
Lugano, Switzerland
14-17 July 2003

ELECTRONIC PROCEEDINGS

Radu Lazar, Glen Meeden

Exploring a Collection of Priors Arising from an Imprecise Probability Assessment Based on Linear Constraints

Abstract

Consider the situation where the available prior informa-tion is only sufficient to identify a class of possible prior dis-tributions. In such cases it would be of interest to be able to explore the behavior of functions defined on this class. Here we develop a method based on the Metropolis-Hastings al-gorithm that allows one to investigate an imprecise prior as-sessment based on linear constraints.

Keywords. linear constraints, probability assessment, Bayesian inference, Metropolis-Hastings algorithm

Paper Download

The paper is availabe in the following formats:

Authors addresses:

Radu Lazar
313 Ford Hall
School of Statistics
University of Minnesota
Minneapolis, MN 55455

Glen Meeden
School of Statistics University of Minnesota
"313 Ford Hall, 224 Church St. SE",
55455-0493,Minneapolis,MN
USA

E-mail addresses:

Radu Lazar lazar@stat.umn.edu
Glen Meeden glen@stat.umn.edu

[ back to the Proceedings of ISIPTA '03 home page ]  
Send any remarks to the following address: smc@decsai.ugr.es