Statistical Papers

, Volume 51, Issue 3, pp 613–628

Eliciting vague but proper maximal entropy priors in Bayesian experiments

Authors

    • Department of Mathematics and StatisticsLaval University
Regular Article

DOI: 10.1007/s00362-008-0149-9

Cite this article as:
Bousquet, N. Stat Papers (2010) 51: 613. doi:10.1007/s00362-008-0149-9

Abstract

Priors elicited according to maximal entropy rules have been used for years in objective and subjective Bayesian analysis. However, when the prior knowledge remains fuzzy or dubious, they often suffer from impropriety which can make them uncomfortable to use. In this article we suggest the formal elicitation of an encompassing family for the standard maximal entropy (ME) priors and the maximal data information (MDI) priors, which can lead to obtain proper families. An interpretation is given in the objective framework of channel coding. In a subjective framework, the performance of the method is shown in a reliability context when flat but proper priors are elicited for the Weibull lifetime distributions. Such priors appear as practical tools for sensitivity studies.

Keywords

Bayesian inferenceExpert opinionKullback–Leibler distanceShannon’s entropyNoninformative priorsChannel codingSensitivity studyWeibull

Mathematics Subject Classification (2000)

65K0590C35

Copyright information

© Springer-Verlag 2008