Using prior expansions for prior-data conflict checking
From MaRDI portal
Abstract: Any Bayesian analysis involves combining information represented through different model components, and when different sources of information are in conflict it is important to detect this. Here we consider checking for prior-data conflict in Bayesian models by expanding the prior used for the analysis into a larger family of priors, and considering a marginal likelihood score statistic for the expansion parameter. Consideration of different expansions can be informative about the nature of any conflict, and extensions to hierarchically specified priors and connections with other approaches to prior-data conflict checking are discussed. Implementation in complex situations is illustrated with two applications. The first concerns testing for the appropriateness of a LASSO penalty in shrinkage estimation of coefficients in linear regression. Our method is compared with a recent suggestion in the literature designed to be powerful against alternatives in the exponential power family, and we use this family as the prior expansion for constructing our check. A second application concerns a problem in quantum state estimation, where a multinomial model is considered with physical constraints on the model parameters. In this example, the usefulness of different prior expansions is demonstrated for obtaining checks which are sensitive to different aspects of the prior.
Recommendations
- Checking for prior-data conflict
- A limit result for the prior predictive applied to checking for prior-data conflict
- Inference after checking multiple Bayesian models for data conflict and applications to mitigating the influence of rejected priors
- Avoiding prior-data conflict in regression models via mixture priors
- Comment: Bayesian checking of the second levels of hierarchical models
Cites work
- A graphical diagnostic for identifying influential model choices in Bayesian hierarchical models
- A limit result for the prior predictive applied to checking for prior-data conflict
- A robust conflict measure of inconsistencies in Bayesian hierarchical models
- A simulation-intensive approach for checking hierarchical models
- Adaptive prior weighting in generalized regression
- Bayesian checking of the second levels of hierarchical models
- Bayesian influence analysis: a geometric approach
- Bayesian revision of a prior given prior-data conflict, expert opinion, or a similar insight: a large-deviation approach
- Checking for prior-data conflict
- Conflict diagnostics in directed acyclic graphs, with applications in Bayesian evidence synthesis
- Decomposing posterior variance
- Diagnostics of prior-data agreement in applied Bayesian analysis
- Extensions of a conflict measure of inconsistencies in Bayesian hierarchical models
- Goodness-of-fit diagnostics for Bayesian hierarchical models
- scientific article; zbMATH DE number 3173999 (Why is no real title available?)
- scientific article; zbMATH DE number 720675 (Why is no real title available?)
- scientific article; zbMATH DE number 947414 (Why is no real title available?)
- scientific article; zbMATH DE number 845714 (Why is no real title available?)
- Hypothesis assessment and inequalities for Bayes factors and relative belief ratios
- Identifying outliers in Bayesian hierarchical models: a simulation-based approach
- Inference in hidden Markov models.
- Invariant \(P\)-values for model checking
- Measuring statistical evidence using relative belief
- Model criticism in latent space
- On the overall sensitivity of the posterior distribution to its inputs
- Optimal error regions for quantum state estimation
- Optimal robustness results for relative belief inferences and the relationship to prior-data conflict
- P Values for Composite Null Models
- Quantum state estimation.
- Sampling and Bayes' Inference in Scientific Modelling and Robustness
- Sensitivity analysis for Bayesian hierarchical models
- Sensitivity in Bayesian Statistics: The Prior and the Likelihood
- The horseshoe estimator for sparse signals
- Weak informativity and the information in one prior relative to another
Cited in
(13)- A necessary Bayesian nonparametric test for assessing multivariate normality
- Post-data evaluation of prior assessment
- Comment: Bayesian checking of the second levels of hierarchical models
- Prior-based model checking
- Avoiding prior-data conflict in regression models via mixture priors
- On one-sample Bayesian tests for the mean
- Adaptive prior weighting in generalized regression
- A limit result for the prior predictive applied to checking for prior-data conflict
- Checking for prior-data conflict
- Quantifying observed prior impact
- Detecting and diagnosing prior and likelihood sensitivity with power-scaling
- Using prior-data conflict to tune Bayesian regularized regression models
- Graphical and summary diagnostics for node level adequacy in Bayesian hierarchical models
This page was built for publication: Using prior expansions for prior-data conflict checking
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2057380)
