Testing hypotheses via a mixture estimation model

Abstract

We consider a novel paradigm for Bayesian testing of hypotheses and Bayesian model comparison. Our alternative to the traditional construction of posterior probabilities that a given hypothesis is true or that the data originates from a specific model is to consider the models under comparison as components of a mixture model. We therefore replace the original testing problem with an estimation one that focus on the probability weight of a given model within a mixture model. We analyze the sensitivity on the resulting posterior distribution on the weights of various prior modeling on the weights. We stress that a major appeal in using this novel perspective is that generic improper priors are acceptable, while not putting convergence in jeopardy. Among other features, this allows for a resolution of the Lindley-Jeffreys paradox. When using a reference Beta B(a,a) prior on the mixture weights, we note that the sensitivity of the posterior estimations of the weights to the choice of a vanishes with the sample size increasing and avocate the default choice a=0.5, derived from Rousseau and Mengersen (2011). Another feature of this easily implemented alternative to the classical Bayesian solution is that the speeds of convergence of the posterior mean of the weight and of the corresponding posterior probability are quite similar.

arXiv comment25 pages, 6 figures, 2 tables

References in corpus (6)

• Misspecification in infinite-dimensional Bayesian statistics
• Properties of Nested Sampling
• Frequentist statistics as a theory of inductive inference
• Importance sampling methods for Bayesian discrimination between embedded models
• Bayesian Inference on Mixtures of Distributions
• The Whetstone and the Alum Block: Balanced Objective Bayesian Comparison of Nested Models for Discrete Data

Cited by in corpus (12)

• E-values: Calibration, combination, and applications
• Bayesian hierarchical stacking: Some models are (somewhere) useful
• Stacking for Non-mixing Bayesian Computations: The Curse and Blessing of Multimodal Posteriors
• Penalising model component complexity: A principled, practical approach to constructing priors
• Bayesian model comparison with the Hyvärinen score: computation and consistency
• A Parsimonious Tour of Bayesian Model Uncertainty
• A discussion of "Bayesian model selection based on proper scoring rules" by A.P. Dawid and M. Musio
• A new Bayesian two-sample t-test for effect size estimation under uncertainty based on a two-component Gaussian mixture with known allocations and the region of practical equivalence
• Rejoinder to "Bayesian Model Selection Based on Proper Scoring Rules"
• Bayes Factors via Savage-Dickey Supermodels
• Computer code validation via mixture model estimation
• Bayesian significance test for discriminating between survival distributions