A discussion of "Bayesian model selection based on proper scoring rules" by A.P. Dawid and M. Musio

TL;DR

This paper discusses the use of proper scoring rules in Bayesian model selection, highlighting potential practical challenges despite its conceptual appeal.

Contribution

It provides a critical discussion of Dawid and Musio's approach, emphasizing possible difficulties in practical implementation.

Findings

Highlights practical difficulties with proper scoring rules in Bayesian model selection

Acknowledges the conceptual value of proper scoring rules including improper priors

Encourages further investigation into practical applications

Abstract

This note is a discussion of the article "Bayesian model selection based on proper scoring rules" by A.P. Dawid and M. Musio, to appear in Bayesian Analysis. While appreciating the concepts behind the use of proper scoring rules, including the inclusion of improper priors, we point out here some possible practical difficulties with the advocated approach.