A note on Bayesian model selection for discrete data using proper scoring rules

DOI10.1016/J.SPL.2017.05.010MaRDI QIDQ1687199zbMATH OpenOpenAlexWikidataFDO

Authors A. Philip Dawid, Monica Musio, Silvia Columbu

Publication date 22 December 2017

Published in Statistics \& Probability Letters (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1703.03353

zbMATH Keywords

consistent model selectiondiscrete datahomogeneous scoreprequential

Mathematics Subject Classification ID

Foundations and philosophical topics in statistics (62A01) Statistical decision theory (62C99) Bayesian inference (62F15)

Abstract: We consider the problem of choosing between parametric models for a discrete observable, taking a Bayesian approach in which the within-model prior distributions are allowed to be improper. In order to avoid the ambiguity in the marginal likelihood function in such a case, we apply a homogeneous scoring rule. For the particular case of distinguishing between Poisson and Negative Binomial models, we conduct simulations that indicate that, applied prequentially, the method will consistently select the true model.

Recommendations

• Bayesian model selection based on proper scoring rules
• Context-dependent score based Bayesian information criteria
• Bayesian Model Selection: Measuring the χ²Discrepancy with the Uniform Distribution
• scientific article; zbMATH DE number 1522708
• Theory and applications of proper scoring rules

Cites work

• Estimation of non-normalized statistical models by score matching
• Proper local scoring rules
• Proper local scoring rules on discrete sample spaces
• Rejoinder: ``Bayesian model selection based on proper scoring rules

Cited in

(7)

• On a class of objective priors from scoring rules (with discussion)
• Objective Bayesian model choice for non-nested families: the case of the Poisson and the negative binomial
• Bayesian model selection based on proper scoring rules
• Context-dependent score based Bayesian information criteria
• Bayesian model comparison with the Hyvärinen score: computation and consistency
• The whetstone and the alum block: balanced objective Bayesian comparison of nested models for discrete data
• The Hyvärinen scoring rule in Gaussian linear time series models

This page was built for publication: A note on Bayesian model selection for discrete data using proper scoring rules

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1687199)