False Discovery Rate Smoothing

DOI10.1080/01621459.2017.1319838zbMATH Open1402.62011arXiv1411.6144OpenAlexW1844989601MaRDI QIDQ4559697FDOQ4559697

Russell A. Poldrack, James G. Scott, Oluwasanmi Koyejo, Wesley Tansey

Publication date: 4 December 2018

Published in: Journal of the American Statistical Association (Search for Journal in Brave)

Abstract: We present false discovery rate smoothing, an empirical-Bayes method for exploiting spatial structure in large multiple-testing problems. FDR smoothing automatically finds spatially localized regions of significant test statistics. It then relaxes the threshold of statistical significance within these regions, and tightens it elsewhere, in a manner that controls the overall false-discovery rate at a given level. This results in increased power and cleaner spatial separation of signals from noise. The approach requires solving a non-standard high-dimensional optimization problem, for which an efficient augmented-Lagrangian algorithm is presented. In simulation studies, FDR smoothing exhibits state-of-the-art performance at modest computational cost. In particular, it is shown to be far more robust than existing methods for spatially dependent multiple testing. We also apply the method to a data set from an fMRI experiment on spatial working memory, where it detects patterns that are much more biologically plausible than those detected by standard FDR-controlling methods. All code for FDR smoothing is publicly available in Python and R.

Full work available at URL: https://arxiv.org/abs/1411.6144

Recommendations

• False discovery rate envelopes
• False Discovery Rate Control via Data Splitting
• On a generalized false discovery rate
• A Direct Approach to False Discovery Rates
• On methods controlling the false discovery rate
• Controlling the false discovery rate via knockoffs
• Correcting false discovery rates for their bias toward false positives
• False discovery rate for functional data
• False discovery rate control for effect modification in observational studies

zbMATH Keywords

false discovery ratemultiple hypothesis testingempirical BayesFDRspatial smoothingfused Lasso

Mathematics Subject Classification ID

Applications of statistics to biology and medical sciences; meta analysis (62P10) Software, source code, etc. for problems pertaining to statistics (62-04) Hypothesis testing in multivariate analysis (62H15) Empirical decision procedures; empirical Bayes procedures (62C12)

Cites Work

• SparseNet: Coordinate Descent With Nonconvex Penalties
• Nonlinear total variation based noise removal algorithms
• The Adaptive Lasso and Its Oracle Properties
• Distributed Optimization and Statistical Learning via the Alternating Direction Method of Multipliers
• Simultaneous analysis of Lasso and Dantzig selector
• A general framework for multiple testing dependence
• Title not available (Why is that?)
• Sparsity and Smoothness Via the Fused Lasso
• The solution path of the generalized lasso
• On total variation minimization and surface evolution using parametric maximum flows
• Robustness of multiple testing procedures against dependence
• Large-scale multiple testing under dependence
• False Discovery Rates for Spatial Signals
• Degrees of freedom in lasso problems
• Bayes and empirical-Bayes multiplicity adjustment in the variable-selection problem
• Large-Scale Simultaneous Hypothesis Testing
• An exploration of aspects of Bayesian multiple testing
• Empirical Bayes Analysis of a Microarray Experiment
• False discovery rate analysis of brain diffusion direction maps
• Handbook of Functional MRI Data Analysis
• False Discovery Control for Random Fields
• False Discovery Control in Large-Scale Spatial Multiple Testing
• Microarrays, empirical Bayes and the two-groups model
• Asymptotic Bayes-optimality under sparsity of some multiple testing procedures
• Testing precise hypotheses. With comments and a rejoinder by the authors
• A Bayesian Mixture Model for Differential Gene Expression
• Title not available (Why is that?)
• A nonparametric empirical Bayes framework for large-scale multiple testing
• Title not available (Why is that?)
• Consistency of a recursive estimate of mixing distributions
• Multiple testing via \(\mathrm{FDR}_L\) for large-scale imaging data
• Title not available (Why is that?)
• Local Shrinkage Rules, Lévy Processes and Regularized Regression
• Multiple testing for neuroimaging via hidden Markov random field

Cited In (18)

• Large-scale multiple testing via multivariate hidden Markov models
• Prior distributions for objective Bayesian analysis
• Discussion of ``Optimal test procedures for multiple hypotheses controlling the familywise expected loss by Willi Maurer, Frank Bretz, and Xiaolei Xun
• Bayesian inference and testing of group differences in brain networks
• Optimal false discovery rate control for large scale multiple testing with auxiliary information
• Detection of Local Differences in Spatial Characteristics Between Two Spatiotemporal Random Fields
• LAWS: A Locally Adaptive Weighting and Screening Approach to Spatial Multiple Testing
• Title not available (Why is that?)
• Flexible and Fast Spatial Return Level Estimation Via a Spatially Fused Penalty
• A survey of nonparametric mixing density estimation via the predictive recursion algorithm
• Testing and estimation for clustered signals
• smoothfdr
• Double Empirical Bayes Testing
• Quantile regression by dyadic CART
• Revisiting consistency of a recursive estimator of mixing distributions
• A PRticle filter algorithm for nonparametric estimation of multivariate mixing distributions
• A spatially adaptive large-scale multiple-testing procedure
• Estimating a mixing distribution on the sphere using predictive recursion

Uses Software

• R
• sparsenet
• glmgen
• GitHub

This page was built for publication: False Discovery Rate Smoothing

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4559697)