Robust inference via multiplier bootstrap

DOI10.1214/19-AOS1863zbMATH Open1458.62075arXiv1903.07208OpenAlexW3043556213MaRDI QIDQ2196240FDOQ2196240

Authors: Yanyan Li 

Publication date: 28 August 2020

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

Abstract: This paper investigates the theoretical underpinnings of two fundamental statistical inference problems, the construction of confidence sets and large-scale simultaneous hypothesis testing, in the presence of heavy-tailed data. With heavy-tailed observation noise, finite sample properties of the least squares-based methods, typified by the sample mean, are suboptimal both theoretically and empirically. In this paper, we demonstrate that the adaptive Huber regression, integrated with the multiplier bootstrap procedure, provides a useful robust alternative to the method of least squares. Our theoretical and empirical results reveal the effectiveness of the proposed method, and highlight the importance of having inference methods that are robust to heavy tailedness.

Full work available at URL: https://arxiv.org/abs/1903.07208

Recommendations

• A new perspective on robust \(M\)-estimation: finite sample theory and applications to dependence-adjusted multiple testing
• Tests of hypotheses about regression parameters when using a robust estimator
• Scale calibration for high-dimensional robust regression
• scientific article; zbMATH DE number 1301884
• Multivariate regression \(S\)-estimators for robust estimation and inference

zbMATH Keywords

multiple testingrobust regressionmultiplier bootstrapconfidence setsheavy-tailed dataHuber loss

Mathematics Subject Classification ID

Bootstrap, jackknife and other resampling methods (62F40) Linear regression; mixed models (62J05) Large deviations (60F10) Robustness and adaptive procedures (parametric inference) (62F35) Paired and multiple comparisons; multiple testing (62J15)

Cites Work

• A factor model approach to multiple testing under dependence
• Weak convergence and empirical processes. With applications to statistics
• Title not available (Why is that?)
• Estimating False Discovery Proportion Under Arbitrary Covariance Dependence
• Robust Estimation of a Location Parameter
• Gaussian approximations and multiplier bootstrap for maxima of sums of high-dimensional random vectors
• High-Dimensional Probability
• Robust Statistics
• Geometric median and robust estimation in Banach spaces
• Challenging the empirical mean and empirical variance: a deviation study
• Loss minimization and parameter estimation with heavy tails
• Common risk factors in the returns on stocks and bonds
• Robust linear least squares regression
• Multiple testing procedures with applications to genomics.
• Large-scale inference. Empirical Bayes methods for estimation, testing, and prediction
• Title not available (Why is that?)
• Robustness and Accuracy of Methods for High Dimensional Data Analysis Based on Student’s t-Statistic
• Anti-concentration and honest, adaptive confidence bands
• Phase transition and regularized bootstrap in large-scale \(t\)-tests with false discovery rate control
• Estimation of High Dimensional Mean Regression in the Absence of Symmetry and Light Tail Assumptions
• To How Many Simultaneous Hypothesis Tests Can Normal, Student'stor Bootstrap Calibration Be Applied?
• Some nonasymptotic results on resampling in high dimension. I: Confidence regions
• Generalized bootstrap for estimating equations
• Robust PCA and pairs of projections in a Hilbert space
• Title not available (Why is that?)
• Sub-Gaussian mean estimators
• Bootstrap confidence sets under model misspecification
• Empirical risk minimization for heavy-tailed losses
• Sub-Gaussian estimators of the mean of a random matrix with heavy-tailed entries
• Confounder adjustment in multiple hypothesis testing
• The asymptotic joint distribution of self-normalized censored sums and sums of squares
• Power Enhancement in High-Dimensional Cross-Sectional Tests
• Sub-Gaussian estimators of the mean of a random vector
• A new perspective on robust \(M\)-estimation: finite sample theory and applications to dependence-adjusted multiple testing
• Cross-Dimensional Inference of Dependent High-Dimensional Data

Cited In (12)

• Volatility prediction comparison via robust volatility proxies: an empirical deviation perspective
• Gaussian differentially private robust mean estimation and inference
• Large-dimensional central limit theorem with fourth-moment error bounds on convex sets and balls
• Multiplier bootstrap methods for conditional distributions
• Ridge regression revisited: debiasing, thresholding and bootstrap
• Robust high-dimensional tuning free multiple testing
• Smoothed quantile regression with large-scale inference
• Robust Signal Recovery for High-Dimensional Linear Log-Contrast Models with Compositional Covariates
• Iteratively reweighted \(\ell_1\)-penalized robust regression
• Robust inference for high‐dimensional single index models
• Title not available (Why is that?)
• Robustness of Bootstrap in Instrumental Variable Regression

Uses Software

• FAMT

This page was built for publication: Robust inference via multiplier bootstrap

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2196240)