Deriving proper uniform priors for regression coefficients, part II
H. R. N. van Erp, R. O. Linger, Pieter H.A.J.M. van Gelder · AIP conference proceedings · 2017
It is a relatively well-known fact that in problems of Bayesian model selection improper priors should, in general, be avoided. In this paper we proceed to derive a class of proper uniform priors for regression models. We then use these priors to derive the corresponding evidence values of the regression models under consideration. These evidences values are then connected to the implied evidence of the Bayesian Information Criterion (BIC).