Bayes Factors for Linear and Log-Linear Models with Vague Prior Information
D. J. Spiegelhalter, A. F. M. Smith
Journal of the Royal Statistical Society: Series B (Statistical Methodology) · 1982 · Volume 44 · Issue 3
DOI: 10.1111/j.2517-6161.1982.tb01217.x
Abstract
The Bayes factor, or integrated likelihood ratio, provides a possible Bayesian alternative to the standardF-test procedure for comparing two linear models. However, if vague prior information for the model parameters is represented by limiting improper prior forms, it is well known that the resulting Bayes factor involves an arbitrary, unspecified constant, and is thus not well defined. A method of assigning this constant is proposed and illustrated for a number of standard problems. The approach is then extended to the analysis of contingency tables using log-linear models.
Keywords
bayes factors, linear models, vague prior information, imaginary observations, one-way anova, randomized complete blocks, regression, contingency tables, log-linear models
Enable JavaScript for citation tools and additional paper interactions.