Bayesian model selection based on proper scoring rules

DOI10.1214/15-BA942MaRDI QIDQ273612zbMATH OpenOpenAlexFDO

Authors Monica Musio, A. Philip Dawid

Publication date 22 April 2016

Published in Bayesian Analysis (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1409.5291, https://projecteuclid.org/euclid.ba/1423083641

zbMATH Keywords

consistent model selectionhomogeneous scoreprequentialHyvärinen score

Mathematics Subject Classification ID

Point estimation (62F10) Bayesian inference (62F15) Bayesian problems; characterization of Bayes procedures (62C10)

Abstract: Bayesian model selection with improper priors is not well-defined because of the dependence of the marginal likelihood on the arbitrary scaling constants of the within-model prior densities. We show how this problem can be evaded by replacing marginal log-likelihood by a homogeneous proper scoring rule, which is insensitive to the scaling constants. Suitably applied, this will typically enable consistent selection of the true model.

Recommendations

• A note on Bayesian model selection for discrete data using proper scoring rules
• Model selection with vague prior information
• Context-dependent score based Bayesian information criteria
• An Intrinsic Limiting Procedure for Model Selection and Hypotheses Testing
• Objective Bayesian inference with proper scoring rules

Cites work

• scientific article; zbMATH DE number 3333814 (Why is no real title available?)
• scientific article; zbMATH DE number 3390151 (Why is no real title available?)
• Bayesian model selection in high-dimensional settings
• Consistency of objective Bayes factors as the model dimension grows
• Criteria for Bayesian model choice with application to variable selection
• Estimation of non-normalized statistical models by score matching
• Estimation of spatial processes using local scoring rules: spatial special issue
• Game theory, maximum entropy, minimum discrepancy and robust Bayesian decision theory
• Minimum scoring rule inference
• ON THE QUESTION OF ABSOLUTE CONTINUITY AND SINGULARITY OF PROBABILITY MEASURES
• Present Position and Potential Developments: Some Personal Views: Statistical Theory: The Prequential Approach
• Proper local scoring rules
• The Intrinsic Bayes Factor for Model Selection and Prediction
• Theory and applications of proper scoring rules

Cited in

(20)

• An Invitation to Sequential Monte Carlo Samplers
• scientific article; zbMATH DE number 7415101 (Why is no real title available?)
• The Hyvärinen scoring rule in Gaussian linear time series models
• Theory and applications of proper scoring rules
• Minimax predictive density for sparse count data
• Adaptation of the tuning parameter in general Bayesian inference with robust divergence
• Objective Bayesian inference with proper scoring rules
• Bayesian quantile regression using the skew exponential power distribution
• Bayesian model comparison with the Hyvärinen score: computation and consistency
• Prior distributions for objective Bayesian analysis
• Bayes factor asymptotics for variable selection in the Gaussian process framework
• The role of Bayesian philosophy within Bayesian model selection
• Bayesian cross-validation by parallel Markov chain Monte Carlo
• An objective Bayes factor with improper priors
• A note on Bayesian model selection for discrete data using proper scoring rules
• Context-dependent score based Bayesian information criteria
• Generalized Bayesian Inference for Discrete Intractable Likelihood
• scientific article; zbMATH DE number 7626771 (Why is no real title available?)
• An integrated surrogate model constructing method: annealing combinable Gaussian process
• Comparison of stochastic frontier models using the Hyvärinen factor

This page was built for publication: Bayesian model selection based on proper scoring rules

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q273612)