On a class of objective priors from scoring rules (with discussion)
From MaRDI portal
Abstract: Objective prior distributions represent an important tool that allows one to have the advantages of using the Bayesian framework even when information about the parameters of a model is not available. The usual objective approaches work off the chosen statistical model and in the majority of cases the resulting prior is improper, which can pose limitations to a practical implementation, even when the complexity of the model is moderate. In this paper we propose to take a novel look at the construction of objective prior distributions, where the connection with a chosen sampling distribution model is removed. We explore the notion of defining objective prior distributions which allow one to have some degree of flexibility, in particular in exhibiting some desirable features, such as being proper, or centered on specific values which would be of interest in nested model comparisons. The basic tool we use are proper scoring rules and the main result is a class of objective prior distributions that can be employed in scenarios where the usual model based priors fail, such as mixture models and model selection via Bayes factors. In addition, we show that the proposed class of priors is the result of minimising the information it contains, providing solid interpretation to the method.
Recommendations
- Objective Bayesian inference with proper scoring rules
- Objective priors: an introduction for frequentists
- Objective priors in the empirical Bayes framework
- An objective Bayesian criterion to determine model prior probabilities
- Objective priors for discrete parameter spaces
- Prior distributions for objective Bayesian analysis
- An objective Bayes perspective on p‐values
- Objective and subjective foundations for multiple priors
- An objective prior for hyperparameters in normal hierarchical models
- An objective Bayes factor with improper priors
Cites work
- A note on Bayesian model selection for discrete data using proper scoring rules
- A note on the existence of the posterior distribution for a class of mixed models for binomial responses
- A universal prior for integers and estimation by minimum description length
- An invariant form for the prior probability in estimation problems
- An objective approach to prior mass functions for discrete parameter spaces
- Bayes factors and marginal distributions in invariant situations
- Bayesian analysis for random coefficient regression models using noninformative priors
- Bounds on the deficit in the logarithmic Sobolev inequality
- Estimation of non-normalized statistical models by score matching
- Flexible linear mixed models with improper priors for longitudinal and survival data
- scientific article; zbMATH DE number 4133286 (Why is no real title available?)
- scientific article; zbMATH DE number 3426675 (Why is no real title available?)
- scientific article; zbMATH DE number 578421 (Why is no real title available?)
- scientific article; zbMATH DE number 720676 (Why is no real title available?)
- scientific article; zbMATH DE number 3189754 (Why is no real title available?)
- scientific article; zbMATH DE number 3190795 (Why is no real title available?)
- Information Theory and Statistical Mechanics
- Invariant Prior Distributions
- Jeffreys priors for mixture estimation: properties and alternatives
- Nonsubjective priors via predictive relative entropy regret
- Objective Bayesian analysis for the Student-t regression model
- Objective Bayesian inference with proper scoring rules
- Objective priors for discrete parameter spaces
- On Bayesian Analysis of Generalized Linear Models Using Jeffreys's Prior
- On the independence Jeffreys prior for skew-symmetric models
- On the use of non-local prior densities in Bayesian hypothesis tests
- Overall objective priors
- Overdispersed generalized linear models
- Penalising model component complexity: a principled, practical approach to constructing priors
- Prior distributions for objective Bayesian analysis
- Prior Probabilities
- Proper local scoring rules
- Pushing the Limits of Contemporary Statistics: Contributions in Honor of Jayanta K. Ghosh
- Rejoinder: ``Bayesian model selection based on proper scoring rules
- Statistical analysis of finite mixture distributions
- Strong Inconsistency from Uniform Priors
- The formal definition of reference priors
- The Intrinsic Bayes Factor for Model Selection and Prediction
- The Selection of Prior Distributions by Formal Rules
- The whetstone and the alum block: balanced objective Bayesian comparison of nested models for discrete data
- Un-Bayesian implications of improper Bayes inference in routine statistical problems
- Variational methods in statistics
Cited in
(11)- Prior distributions for objective Bayesian analysis
- On a prior based on the Wasserstein information matrix
- Objective Bayesian inference with proper scoring rules
- Preposterior expected loss as a scoring rule for prior distributions
- Discussion of ``Objective priors: An introduction for frequentists by M. Ghosh
- Objective priors for discrete parameter spaces
- An objective approach to prior mass functions for discrete parameter spaces
- Prior-free inference for objective Bayesian analysis and model selection
- Quantifying observed prior impact
- An objective non-local prior for skew-symmetric models
- An objective prior from cumulative probabilities for censored data
This page was built for publication: On a class of objective priors from scoring rules (with discussion)
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2057364)
