Optional stopping with Bayes factors: a categorization and extension of folklore results, with an application to invariant situations
From MaRDI portal
Publication:6201435
DOI10.1214/20-BA1234arXiv1807.09077MaRDI QIDQ6201435FDOQ6201435
Authors: Allard Hendriksen, Rianne de Heide, Peter D. Grünwald 
Publication date: 20 February 2024
Published in: Bayesian Analysis (Search for Journal in Brave)
Abstract: It is often claimed that Bayesian methods, in particular Bayes factor methods for hypothesis testing, can deal with optional stopping. We first give an overview, using elementary probability theory, of three different mathematical meanings that various authors give to this claim: (1) stopping rule independence, (2) posterior calibration and (3) (semi-) frequentist robustness to optional stopping. We then prove theorems to the effect that these claims do indeed hold in a general measure-theoretic setting. For claims of type (2) and (3), such results are new. By allowing for non-integrable measures based on improper priors, we obtain particularly strong results for the practically important case of models with nuisance parameters satisfying a group invariance (such as location or scale). We also discuss the practical relevance of (1)--(3), and conclude that whether Bayes factor methods actually perform well under optional stopping crucially depends on details of models, priors and the goal of the analysis.
Full work available at URL: https://arxiv.org/abs/1807.09077
Cites Work
- On confidence sequences
- A STATISTICAL PARADOX
- Mixtures of g Priors for Bayesian Variable Selection
- Title not available (Why is that?)
- Almost the Best of Three Worlds: Risk, Consistency and Optional Stopping for the Switch Criterion in Nested Model Selection
- Title not available (Why is that?)
- Title not available (Why is that?)
- Test martingales, Bayes factors and \(p\)-values
- Default Bayes factors for ANOVA designs
- Distributions of maximal invariants using quotient measures
- The case for objective Bayesian analysis
- Criteria for Bayesian model choice with application to variable selection
- Objective priors for the bivariate normal model
- Title not available (Why is that?)
- The minimum description length principle in coding and modeling
- Title not available (Why is that?)
- Bayes factors and marginal distributions in invariant situations
- Rejection odds and rejection ratios: a proposal for statistical practice in testing hypotheses
- Unified Conditional Frequentist and Bayesian Testing of Composite Hypotheses
- Title not available (Why is that?)
- Statistical monitoring of clinical trials. A unified approach.
- Simulation-Based Approaches for Verification of Embedded Control Systems: An Overview of Traditional and Advanced Modeling, Testing, and Verification Techniques
- Group invariant inference and right Haar measure
- Safe Probability: Restricted Conditioning and Extended Marginalization
- Time-uniform Chernoff bounds via nonnegative supermartingales
Cited In (2)
This page was built for publication: Optional stopping with Bayes factors: a categorization and extension of folklore results, with an application to invariant situations
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6201435)
