Combining expert opinions in prior elicitation

Isabelle I. Albert (1) , Sophie S. Donnet (2) , Chantal C. Guihenneuc-Jouyaux (2) , Samantha S. Low-Choy , Kerrie K. Mengersen , Judith J. Rousseau (3)

1 MET@RISK - Méthodologies d'Analyse de Risque Alimentaire
2 CREST - Centre de Recherche en Économie et Statistique
3 CEREMADE - CEntre de REcherches en MAthématiques de la DEcision

We consider the problem of combining opinions from different experts in an explicitly model-based way to construct a valid subjective prior in a Bayesian statistical approach. We propose a generic approach by considering a hierarchical model accounting for various sources of variation as well as accounting for potential dependence between experts. We apply this approach to two problems. The first problem deals with a food risk assessment problem involving modelling dose-response for Listeria monocytogenes contamination of mice. Two hierarchical levels of variation are considered (between and within experts) with a complex mathematical situation due to the use of an indirect probit regression. The second concerns the time taken by PhD students to submit their thesis in a particular school. It illustrates a complex situation where three hierarchical levels of variation are modelled but with a simpler underlying probability distribution (log-Normal).

Mots clés: bayesian statistics; hierarchical model; random effects; risk assessment

Domaines: Calcul [stat.CO]

Isabelle I. Albert, Sophie S. Donnet, Chantal C. Guihenneuc-Jouyaux, Samantha S. Low-Choy, Kerrie K. Mengersen, et al.. Combining expert opinions in prior elicitation. Bayesian Analysis, 2012, 7 (3), pp.503-531. ⟨10.1214/12-BA717⟩. ⟨hal-01004440⟩