The Open UniversitySkip to content

Elicitation of prior distributions of variable-selection problems in regression

Garthwaite, Paul H. and Dickey, James M. (1992). Elicitation of prior distributions of variable-selection problems in regression. Annals of Statistics, 20(4) pp. 1697–1719.

Full text available as:
PDF (Version of Record) - Requires a PDF viewer such as GSview, Xpdf or Adobe Acrobat Reader
Download (2141Kb) | Preview
DOI (Digital Object Identifier) Link:
Google Scholar: Look up in Google Scholar


This paper addresses the problem of quantifying expert opinion about a normal linear regression model when there is uncertainty as to which independent variables should be included in the model. Opinion is modeled as a mixture of natural conjugate prior distributions with each distribution in the mixture corresponding to a different subset of the independent variables. It is shown that for certain values of the independent variables, the predictive distribution of the dependent variable simplifies from a mixture of $t$-distributions to a single $t$-distribution. Using this result, a method of eliciting the conjugate distributions of the mixture is developed. The method is illustrated in an example.

Item Type: Journal Article
Copyright Holders: 1992 Annals of Statistics
ISSN: 0090-5364
Keywords: probability assessment methods; probability elicitation; prior distribution; variable selection; linear regression
Academic Unit/Department: Mathematics, Computing and Technology > Mathematics and Statistics
Mathematics, Computing and Technology
Item ID: 17957
Depositing User: Sara Griffin
Date Deposited: 21 Aug 2009 13:48
Last Modified: 24 Feb 2016 13:01
Share this page:


Scopus Citations

Actions (login may be required)

Policies | Disclaimer

© The Open University   + 44 (0)870 333 4340