Assigning a value to a power likelihood in a general Bayesian model

DOI10.1093/BIOMET/ASX010zbMATH Open1506.62264arXiv1701.08515OpenAlexW2962940259MaRDI QIDQ5384489FDOQ5384489

Authors: Chris Holmes, Stephen G. Walker 

Publication date: 24 June 2019

Published in: Biometrika (Search for Journal in Brave)

Abstract: Bayesian approaches to data analysis and machine learning are widespread and popular as they provide intuitive yet rigorous axioms for learning from data; see Bernardo and Smith (2004) and Bishop (2006). However, this rigour comes with a caveat that the Bayesian model is a precise reflection of Nature. There has been a recent trend to address potential model misspecification by raising the likelihood function to a power, primarily for robustness reasons, though not exclusively. In this paper we provide a coherent specification of the power parameter once the Bayesian model has been specified in the absence of a perfect model.

Full work available at URL: https://arxiv.org/abs/1701.08515

Recommendations

• Robust likelihood functions in Bayesian inference
• Likelihood robustness
• Bayesian likelihood robustness in linear models
• Bayesian robustness
• scientific article; zbMATH DE number 850164

zbMATH Keywords

exponential familyscore functionFisher informationFisher information distance

Mathematics Subject Classification ID

Bayesian inference (62F15) Applications of statistics to biology and medical sciences; meta analysis (62P10) Robustness and adaptive procedures (parametric inference) (62F35)

Cited In (19)

• Bayesian fractional posteriors
• An Invitation to Sequential Monte Carlo Samplers
• Learning attribute patterns in high-dimensional structured latent attribute models
• Gibbs posterior inference on value-at-risk
• Bayesian model comparison with the Hyvärinen score: computation and consistency
• Universal inference
• Generalized Bayes Quantification Learning under Dataset Shift
• Generalized Bayesian likelihood-free inference
• A comparison of learning rate selection methods in generalized Bayesian inference
• Gibbs posterior inference on multivariate quantiles
• Quantification of empirical determinacy: The impact of likelihood weighting on posterior location and spread in Bayesian meta-analysis estimated with JAGS and INLA
• General Robust Bayes Pseudo-Posteriors: Exponential Convergence Results with Applications
• Model-free posterior inference on the area under the receiver operating characteristic curve
• Empirical priors and coverage of posterior credible sets in a sparse normal mean model
• Empirical priors and posterior concentration in a piecewise polynomial sequence model
• Gibbs posterior concentration rates under sub-exponential type losses
• Bernstein-von Mises theorem and misspecified models: a review
• Observational nonidentifiability, generalized likelihood and free energy
• Empirical priors for prediction in sparse high-dimensional linear regression

This page was built for publication: Assigning a value to a power likelihood in a general Bayesian model

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5384489)