Estimating minimum effect with outlier selection

DOI10.1214/20-AOS1956zbMATH Open1461.62052arXiv1809.08330OpenAlexW2892349150MaRDI QIDQ2656596FDOQ2656596

Authors: Alexandra Carpentier, Étienne Roquain, Nicolas Verzelen, Sylvain Delattre 

Publication date: 11 March 2021

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

Abstract: We introduce one-sided versions of Huber's contamination model, in which corrupted samples tend to take larger values than uncorrupted ones. Two intertwined problems are addressed: estimation of the mean of uncorrupted samples (minimum effect) and selection of corrupted samples (outliers). Regarding the minimum effect estimation, we derive the minimax risks and introduce adaptive estimators to the unknown number of contaminations. Interestingly, the optimal convergence rate highly differs from that in classical Huber's contamination model. Also, our analysis uncovers the effect of particular structural assumptions on the distribution of the contaminated samples. As for the problem of selecting the outliers, we formulate the problem in a multiple testing framework for which the location/scaling of the null hypotheses are unknown. We rigorously prove how estimating the null hypothesis is possible while maintaining a theoretical guarantee on the amount of the falsely selected outliers, both through false discovery rate (FDR) or post hoc bounds. As a by-product, we address a long-standing open issue on FDR control under equi-correlation, which reinforces the interest of removing dependency when making multiple testing.

Full work available at URL: https://arxiv.org/abs/1809.08330

Recommendations

• On Huber's contaminated model
• Confidence regions and minimax rates in outlier-robust estimation on the probability simplex
• Density estimation with contamination: minimax rates and theory of adaptation
• Excess-risk consistency of group-hard thresholding estimator in robust estimation of Gaussian mean
• ERM and RERM are optimal estimators for regression problems when malicious outliers corrupt the labels

zbMATH Keywords

false discovery ratemultiple testingsparsityselective inferenceHermite polynomialsminimax ratemoment matchingcontaminationequicorrelationpost hoc

Mathematics Subject Classification ID

Nonparametric hypothesis testing (62G10) Minimax procedures in statistical decision theory (62C20) Paired and multiple comparisons; multiple testing (62J15)

Cites Work

• A factor model approach to multiple testing under dependence
• The incidental parameter problem since 1948
• A general framework for multiple testing dependence
• Title not available (Why is that?)
• Estimating false discovery proportion under arbitrary covariance dependence
• Robust Estimation of a Location Parameter
• Consistent Estimates Based on Partially Consistent Observations
• Adaptive hypothesis testing using wavelets
• The control of the false discovery rate in multiple testing under dependency.
• Higher criticism for detecting sparse heterogeneous mixtures.
• A stochastic process approach to false discovery control.
• Proportion of Non-Zero Normal Means: Universal Oracle Equivalences and Uniformly Consistent Estimators
• Title not available (Why is that?)
• Estimating the Null and the Proportion of Nonnull Effects in Large-Scale Multiple Comparisons
• Correlation and Large-Scale Simultaneous Significance Testing
• Large-Scale Simultaneous Hypothesis Testing
• An adaptive step-down procedure with proven FDR control under independence
• Exceedance Control of the False Discovery Proportion
• Optimal weighting for false discovery rate control
• Controlling the false discovery rate via knockoffs
• Control of the false discovery rate under dependence using the bootstrap and subsampling
• Control of generalized error rates in multiple testing
• Nonparametric goodness-of-fit testing under Gaussian models
• Non-asymptotic minimax rates of testing in signal detection
• Minimax estimation of linear and quadratic functionals on sparsity classes
• Dependency and false discovery rate: asymptotics
• Multiple testing procedures with applications to genomics.
• Multiple testing with the structure-adaptive Benjamini-Hochberg algorithm
• Title not available (Why is that?)
• Estimation of the false discovery proportion with unknown dependence
• SLOPE-adaptive variable selection via convex optimization
• Controlling the number of false discoveries: application to high-dimensional genomic data
• Exact and Approximate Stepdown Methods for Multiple Hypothesis Testing
• On nonparametric tests of positivity/monotonicity/convexity
• Minimax risks for sparse regressions: ultra-high dimensional phenomenons
• Optimal rates of convergence for estimating the null density and proportion of nonnull effects in large-scale multiple testing
• Bandwidth selection in kernel density estimation: oracle inequalities and adaptive minimax optimality
• New procedures controlling the false discovery proportion via Romano-Wolf's heuristic
• Further results on controlling the false discovery proportion
• Minimal penalty for Goldenshluger-Lepski method
• On Nonparametric Estimation of the Value of a Linear Functional in Gaussian White Noise
• Testing composite hypotheses, Hermite polynomials and optimal estimation of a nonsmooth functional
• On estimation of the \(L_r\) norm of a regression function
• Empirical Bayes estimates for large-scale prediction problems
• Adaptive and minimax optimal estimation of the tail coefficient
• On the false discovery proportion convergence under Gaussian equi-correlation
• Rejoinder to: On methods controlling the false discovery rate
• Distribution-free multiple testing
• Minimax quadratic estimation of a quadratic functional
• Nonquadratic estimators of a quadratic functional
• Doing thousands of hypothesis tests at the same time
• Robust covariance and scatter matrix estimation under Huber's contamination model
• Methodology in robust and nonparametrics statistics.
• Adaptive estimation of the sparsity in the Gaussian vector model
• Optimal adaptive estimation of linear functionals under sparsity
• Multiple testing for exploratory research
• Rejoinder
• On empirical distribution function of high-dimensional Gaussian vector components with an application to multiple testing
• Chebyshev polynomials, moment matching, and optimal estimation of the unseen
• Post hoc confidence bounds on false positives using reference families
• Optimal rates and trade-offs in multiple testing
• On estimation of nonsmooth functionals of sparse normal means

Cited In (6)

• False discovery rate control with unknown null distribution: is it possible to mimic the oracle?
• An inverse Laplace transform oracle estimator for the normal means problem
• Confidence regions and minimax rates in outlier-robust estimation on the probability simplex
• An Empirical Bayes Approach to Controlling the False Discovery Exceedance
• Multidimensional linear functional estimation in sparse Gaussian models and robust estimation of the mean
• Semi-supervised multiple testing

Uses Software

• FAMT

This page was built for publication: Estimating minimum effect with outlier selection

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2656596)