On false discovery rate thresholding for classification under sparsity

DOI10.1214/12-AOS1042zbMATH Open1373.62315arXiv1106.6147OpenAlexW3106252482MaRDI QIDQ741797FDOQ741797

Authors: Pierre Neuvial, Étienne Roquain 

Publication date: 15 September 2014

Published in: The Annals of Statistics (Search for Journal in Brave)

Abstract: We study the properties of false discovery rate (FDR) thresholding, viewed as a classification procedure. The "0"-class (null) is assumed to have a known density while the "1"-class (alternative) is obtained from the "0"-class either by translation or by scaling. Furthermore, the "1"-class is assumed to have a small number of elements w.r.t. the "0"-class (sparsity). We focus on densities of the Subbotin family, including Gaussian and Laplace models. Nonasymptotic oracle inequalities are derived for the excess risk of FDR thresholding. These inequalities lead to explicit rates of convergence of the excess risk to zero, as the number m of items to be classified tends to infinity and in a regime where the power of the Bayes rule is away from 0 and 1. Moreover, these theoretical investigations suggest an explicit choice for the target level $alph{a}_m$ of FDR thresholding, as a function of m. Our oracle inequalities show theoretically that the resulting FDR thresholding adapts to the unknown sparsity regime contained in the data. This property is illustrated with numerical experiments.

Full work available at URL: https://arxiv.org/abs/1106.6147

Recommendations

• Asymptotic minimaxity of false discovery rate thresholding for sparse exponential data
• Adapting to unknown sparsity by controlling the false discovery rate
• Learning false discovery rates by fitting sigmoidal threshold functions
• Adaptive threshold-based classification of sparse high-dimensional data
• False Discovery and its Control in Low Rank Estimation
• The false discovery rate for statistical pattern recognition
• Binary Credal Classification Under Sparsity Constraints
• Sparse linear discriminant analysis by thresholding for high dimensional data
• False discovery rate control via debiased Lasso
• Estimating the false discovery rate using the stochastic approximation algorithm

zbMATH Keywords

classificationfalse discovery ratemultiple testingsparsityadaptive procedureoracle inequalityBayes' rule

Mathematics Subject Classification ID

Bayesian inference (62F15) Classification and discrimination; cluster analysis (statistical aspects) (62H30) Hypothesis testing in multivariate analysis (62H15) Paired and multiple comparisons; multiple testing (62J15)

Cites Work

• Title not available (Why is that?)
• Higher criticism for detecting sparse heterogeneous mixtures.
• A stochastic process approach to false discovery control.
• Asymptotic minimaxity of false discovery rate thresholding for sparse exponential data
• Adapting to unknown sparsity by controlling the false discovery rate
• Adaptive linear step-up procedures that control the false discovery rate
• A Direct Approach to False Discovery Rates
• On the false discovery rate and an asymptotically optimal rejection curve
• An adaptive step-down procedure with proven FDR control under independence
• Exact calculations for false discovery proportion with application to least favorable configura\-tions
• Adaptive false discovery rate control under independence and dependence
• Empirical Bayes Analysis of a Microarray Experiment
• On the Benjamini-Hochberg method
• Some results on false discovery rate in stepwise multiple testing procedures.
• Title not available (Why is that?)
• The positive false discovery rate: A Bayesian interpretation and the \(q\)-value
• Concentration inequalities and model selection. Ecole d'Eté de Probabilités de Saint-Flour XXXIII -- 2003.
• Microarrays, empirical Bayes and the two-groups model
• Asymptotic Bayes-optimality under sparsity of some multiple testing procedures
• Semi-supervised novelty detection
• Operating Characteristics and Extensions of the False Discovery Rate Procedure
• A general decision theoretic formulation of procedures controlling FDR and FNR from a Bayesian perspective
• Some remarks on Simes-type multiple tests of significance.
• Multiple hypotheses testing and expected number of type I errors
• On methods controlling the false discovery rate
• A generalized step-up-down multiple test procedure
• On the performance of FDR control: constraints and a partial solution
• Type I error rate control for testing many hypotheses: a survey with proofs
• On false discovery rate thresholding for classification under sparsity

Cited In (19)

• On spike and slab empirical Bayes multiple testing
• Inferring the finest pattern of mutual independence from data
• Bayes multiple decision functions
• Binary classification with pFDR-pFNR losses
• Asymptotic minimaxity of false discovery rate thresholding for sparse exponential data
• Adapting to unknown sparsity by controlling the false discovery rate
• The Indian official statistical system revisited
• Feature selection by higher criticism thresholding achieves the optimal phase diagram
• Sharp multiple testing boundary for sparse sequences
• On false discovery rate thresholding for classification under sparsity
• On the asymptotic properties of SLOPE
• Optimal classification in sparse Gaussian graphic model
• On the power of some sequential multiple testing procedures
• New procedures controlling the false discovery proportion via Romano-Wolf's heuristic
• Variable selection with Hamming loss
• Rate of convergence of a risk estimator to the normal law in a multiple hypothesis testing problem using the FDR threshold
• Asymptotically optimal sequential FDR and pFDR control with (or without) prior information on the number of signals
• Some optimality properties of FDR controlling rules under sparsity
• Group SLOPE – Adaptive Selection of Groups of Predictors

This page was built for publication: On false discovery rate thresholding for classification under sparsity

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q741797)