Large-scale multiple hypothesis testing with the normal-beta prime prior

DOI10.1080/02331888.2019.1662017zbMATH Open1435.62267arXiv1807.02421OpenAlexW3098182499MaRDI QIDQ5205847FDOQ5205847

Authors: Ray Bai, Malay Ghosh 

Publication date: 17 December 2019

Published in: Statistics (Search for Journal in Brave)

Abstract: We revisit the problem of simultaneously testing the means of $n$ independent normal observations under sparsity. We take a Bayesian approach to this problem by introducing a scale-mixture prior known as the normal-beta prime (NBP) prior. We first derive new concentration properties when the beta prime density is employed for a scale parameter in Bayesian hierarchical models. To detect signals in our data, we then propose a hypothesis test based on thresholding the posterior shrinkage weight under the NBP prior. Taking the loss function to be the expected number of misclassified tests, we show that our test procedure asymptotically attains the optimal Bayes risk when the signal proportion $p$ is known. When $p$ is unknown, we introduce an empirical Bayes variant of our test which also asymptotically attains the Bayes Oracle risk in the entire range of sparsity parameters $ppropto{n}^{-epsilon},epsilonin(0,1)$. Finally, we also consider restricted marginal maximum likelihood (REML) and hierarchical Bayes approaches for estimating a key hyperparameter in the NBP prior and examine multiple testing under these frameworks.

Full work available at URL: https://arxiv.org/abs/1807.02421

Recommendations

• Asymptotic properties of Bayes risk of a general class of shrinkage priors in multiple hypothesis testing under sparsity
• Asymptotic Bayes-optimality under sparsity of some multiple testing procedures
• Bayesian multiple testing under sparsity for polynomial-tailed distributions
• Frequentist properties of Bayesian multiplicity control for multiple testing of normal means
• On optimality of Bayesian testimation in the normal means problem

zbMATH Keywords

multiple testingsparsityempirical Bayesshrinkage priorBayes oracle

Mathematics Subject Classification ID

Ridge regression; shrinkage estimators (Lasso) (62J07) Paired and multiple comparisons; multiple testing (62J15)

Cites Work

• The horseshoe estimator for sparse signals
• The Bayesian Lasso
• Title not available (Why is that?)
• The horseshoe+ estimator of ultra-sparse signals
• Needles and straw in haystacks: Empirical Bayes estimates of possibly sparse sequences
• On the half-Cauchy prior for a global scale parameter
• Dirichlet-Laplace priors for optimal shrinkage
• Some priors for sparse regression modelling
• The future of indirect evidence
• The horseshoe estimator: posterior concentration around nearly black vectors
• Asymptotic Bayes-optimality under sparsity of some multiple testing procedures
• Conditions for posterior contraction in the sparse normal means problem
• Generalized double Pareto shrinkage
• Asymptotic properties of Bayes risk for the horseshoe prior
• Proper Bayes Minimax Estimators of the Multivariate Normal Mean
• A robust generalized Bayes estimator and confidence region for a multivariate normal mean
• Asymptotic properties of Bayes risk of a general class of shrinkage priors in multiple hypothesis testing under sparsity
• Asymptotic optimality of one-group shrinkage priors in sparse high-dimensional problems
• Adaptive posterior contraction rates for the horseshoe
• On spike and slab empirical Bayes multiple testing

Cited In (7)

• Intuitive joint priors for Bayesian linear multilevel models: the R2D2M2 prior
• Bayesian Regression Using a Prior on the Model Fit: The R2-D2 Shrinkage Prior
• Bayesian spatial and spatiotemporal models based on multiscale factorizations
• On the beta prime prior for scale parameters in high-dimensional Bayesian regression models
• Good, great, or lucky? Screening for firms with sustained superior performance using heavy-tailed priors
• Bayes oracle property of multiple tests of multivariate normal means under sparsity
• Asymptotic properties of Bayes risk of a general class of shrinkage priors in multiple hypothesis testing under sparsity

Uses Software

• EBayesThresh

This page was built for publication: Large-scale multiple hypothesis testing with the normal-beta prime prior

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5205847)