A robust Bayesian analysis of variable selection under prior ignorance

DOI10.1007/S13171-022-00287-2MaRDI QIDQ6133754zbMATH OpenOpenAlexFDO

Authors Tathagata Basu, Matthias C. M. Troffaes, Jochen Einbeck

Publication date 21 August 2023

Published in Sankhyā. Series A (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2204.13341

zbMATH Keywords

variable selectionBayesian analysishigh dimensional dataimprecise probability

Mathematics Subject Classification ID

Bayesian inference (62F15) Linear regression; mixed models (62J05)

Abstract: We propose a cautious Bayesian variable selection routine by investigating the sensitivity of a hierarchical model, where the regression coefficients are specified by spike and slab priors. We exploit the use of latent variables to understand the importance of the co-variates. These latent variables also allow us to obtain the size of the model space which is an important aspect of high dimensional problems. In our approach, instead of fixing a single prior, we adopt a specific type of robust Bayesian analysis, where we consider a set of priors within the same parametric family to specify the selection probabilities of these latent variables. We achieve that by considering a set of expected prior selection probabilities, which allows us to perform a sensitivity analysis to understand the effect of prior elicitation on the variable selection. The sensitivity analysis provides us sets of posteriors for the regression coefficients as well as the selection indicators and we show that the posterior odds of the model selection probabilities are monotone with respect to the prior expectations of the selection probabilities. We also analyse synthetic and real life datasets to illustrate our cautious variable selection method and compare it with other well known methods.

Cites work

• scientific article; zbMATH DE number 845714 (Why is no real title available?)
• Bayesian variable selection with shrinking and diffusing priors
• Calibration and empirical Bayes variable selection
• Dirichlet-Laplace priors for optimal shrinkage
• Least angle regression. (With discussion)
• Nearly unbiased variable selection under minimax concave penalty
• On Bayesian lasso variable selection and the specification of the shrinkage parameter
• Optimal predictive model selection.
• Robust Bayesian analysis of the linear regression model
• Robust Bayesian analysis: sensitivity to the prior
• Sampling-Based Approaches to Calculating Marginal Densities
• Spike and slab variable selection: frequentist and Bayesian strategies
• The Adaptive Lasso and Its Oracle Properties
• The Bayesian Lasso
• The imprecise logit-normal model and its application to estimating hazard functions
• Variable Selection via Nonconcave Penalized Likelihood and its Oracle Properties

Cited in

(2)

• Robust Bayesian causal estimation for causal inference in medical diagnosis
• A review of Bayesian variable selection methods: what, how and which

This page was built for publication: A robust Bayesian analysis of variable selection under prior ignorance

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6133754)