Testing order constraints: qualitative differences between Bayes factors and normalized maximum likelihood

DOI10.1016/j.spl.2015.06.014zbMath1396.62042arXiv1411.2778OpenAlexW1761299955MaRDI QIDQ894591

Daniel W. Heck, Richard D. Morey, Eric‐Jan Wagenmakers

Publication date: 1 December 2015

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

Full work available at URL: https://arxiv.org/abs/1411.2778

zbMATH Keywords

model selectionminimum description lengthinequality constraintmodel complexity

Mathematics Subject Classification ID

Bayesian inference (62F15)

Related Items (4)

An evaluation of alternative methods for testing hypotheses, from the perspective of Harold Jeffreys ⋮ Bayes factors, relations to minimum description length, and overlapping model classes ⋮ Selecting amongst multinomial models: an apologia for normalized maximum likelihood ⋮ Multinomial models with linear inequality constraints: overview and improvements of computational methods for Bayesian inference

Cites Work

• The Bayes factor for inequality and about equality constrained models
• Modeling by shortest data description
• Estimating Bayes Factors via Posterior Simulation With the Laplace-Metropolis Estimator
• Strong optimality of the normalized ML models as universal codes and information in data
• Bayes Factors
• Fisher information and stochastic complexity

This page was built for publication: Testing order constraints: qualitative differences between Bayes factors and normalized maximum likelihood