The formal definition of reference priors

Abstract

Reference analysis produces objective Bayesian inference, in the sense that inferential statements depend only on the assumed model and the available data, and the prior distribution used to make an inference is least informative in a certain information-theoretic sense. Reference priors have been rigorously defined in specific contexts and heuristically defined in general, but a rigorous general definition has been lacking. We produce a rigorous general definition here and then show how an explicit expression for the reference prior can be obtained under very weak regularity conditions. The explicit expression can be used to derive new reference priors both analytically and numerically.

Published in at http://dx.doi.org/10.1214/07-AOS587 the Annals of Statistics (http://www.imstat.org/aos/) by the Institute of Mathematical Statistics (http://www.imstat.org)

Cited by in corpus (58)

• Inferential models: A framework for prior-free posterior probabilistic inference
• Overall Objective Priors
• Objective Priors: An Introduction for Frequentists
• BANYAN. IX. The Initial Mass Function and Planetary-Mass Object Space Density of the TW Hya Association
• Marginal inferential models: prior-free probabilistic inference on interest parameters
• Conditional inferential models: combining information for prior-free probabilistic inference
• A search for ttbar resonances in the lepton plus jets final state with ATLAS using 4.7 fb^-1 of pp collisions at sqrt(s) = 7 TeV
• Quantified naturalness from Bayesian statistics
• Objective Bayesian analysis of neutrino masses and hierarchy
• Improving Orbit Estimates for Incomplete Orbits with a New Approach to Priors -- with Applications from Black Holes to Planets
• On nonnegative unbiased estimators
• Functional Uniform Priors for Nonlinear Modelling
• Reference priors for high energy physics
• Neutrino Masses and Mass Hierarchy: Evidence for the Normal Hierarchy
• On an inferential model construction using generalized associations
• A note on p-values interpreted as plausibilities
• Objective Bayesian inference with proper scoring rules
• Macroscopicity of quantum mechanical superposition tests via hypothesis falsification
• Search for contact interactions using the inclusive jet pT spectrum in pp collisions at sqrt(s) = 7 TeV
• On the safe use of prior densities for Bayesian model selection
• WEIRD: Wide-orbit Exoplanet search with InfraRed Direct imaging
• Frameworks for prior-free posterior probabilistic inference
• Using the Data Agreement Criterion to Rank Experts' Beliefs
• Maximum likelihood estimates of diffusion coefficients from single-particle tracking experiments
• A Study on the Power Parameter in Power Prior Bayesian Analysis
• Constraints on photon mass and dark photon from the Jovian magnetic field
• Variational Predictive Information Bottleneck
• Generalized probabilistic principal component analysis of correlated data
• Estimating the selection efficiency
• Conditional probability and improper priors
• Conditional fiducial models
• A frequentist framework of inductive reasoning
• Approximation of improper priors
• Decomposing information into copying versus transformation
• Reference prior for Bayesian estimation of seismic fragility curves
• Asymptotic efficiency of inferential models and a possibilistic Bernstein--von Mises theorem
• Reference analysis of the signal + background model in counting experiments II. Approximate reference prior
• Models of radial velocities and transit light curves
• Comment on "Harold Jeffreys's Theory of Probability Revisited"
• Transitional annealed adaptive slice sampling for Gaussian process hyper-parameter estimation
• Bayesian Quantile Estimation and Regression with Martingale Posteriors
• Large-sample theory for inferential models: a possibilistic Bernstein--von Mises theorem
• Discussion of "On the Birnbaum Argument for the Strong Likelihood Principle"
• Maximum Probability Theorem: A Framework for Probabilistic Learning
• Least-Informative Priors for $0\nu \beta \beta$ Decay Searches
• A unified nonparametric fiducial approach to interval-censored data
• Improper vs finitely additive distributions as limits of countably additive probabilities
• Perspective from the Literature on the Role of Expert Judgment in Scientific and Statistical Research and Practice
• Objective Bayesian meta-analysis based on generalized multivariate random effects model
• Estimation theory and gravity
• A default prior for regression coefficients
• Calibration of the Pareto and related distributions -a reference-intrinsic approach
• Designing Test Information and Test Information in Design
• Comment on Article by Berger, Bernardo, and Sun
• Comment on Article by Berger, Bernardo, and Sun
• Discussion of "Objective Priors: An Introduction for Frequentists" by M. Ghosh
• Discussion of "Objective Priors: An Introduction for Frequentists" by M. Ghosh
• Power-Expected-Posterior Priors as Mixtures of g-Priors