Robust and efficient mean estimation: an approach based on the properties of self-normalized sums

DOI10.1214/21-EJS1925MaRDI QIDQ2074319zbMATH OpenOpenAlexFDO

Authors Stanislav Minsker, Mohamed Ndaoud

Publication date 9 February 2022

Published in Electronic Journal of Statistics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2006.01986, https://projecteuclid.org/journals/electronic-journal-of-statistics/volume-15/issue-2/Robust-and-efficient-mean-estimation-an-approach-based-on/10.1214/21-EJS1925.full

zbMATH Keywords

efficiencyrobust estimationself-normalized sumssub-Gaussian deviations

Mathematics Subject Classification ID

Sampling theory, sample surveys (62D05) Nonparametric robustness (62G35)

Abstract: Let $X$ be a random variable with unknown mean and finite variance. We present a new estimator of the mean of $X$ that is robust with respect to the possible presence of outliers in the sample, provides tight sub-Gaussian deviation guarantees without any additional assumptions on the shape or tails of the distribution, and moreover is asymptotically efficient. This is the first estimator that provably combines all these qualities in one package. Our construction is inspired by robustness properties possessed by the self-normalized sums. Theoretical findings are supplemented by numerical simulations highlighting strong performance of the proposed estimator in comparison with previously known techniques.

Recommendations

• Robust multivariate mean estimation: the optimality of trimmed mean
• Sub-Gaussian estimators of the mean of a random vector
• Robust modifications of U-statistics and applications to covariance estimation problems
• Robust sub-Gaussian estimation of a mean vector in nearly linear time
• Sub-Gaussian mean estimators

Cites work

• A Bound on Tail Probabilities for Quadratic Forms in Independent Random Variables
• A general decision theory for Huber's \(\epsilon\)-contamination model
• Asymptotic Statistics
• Challenging the empirical mean and empirical variance: a deviation study
• Distributed statistical estimation and rates of convergence in normal approximation
• High-dimensional probability. An introduction with applications in data science
• scientific article; zbMATH DE number 3790208 (Why is no real title available?)
• scientific article; zbMATH DE number 1256715 (Why is no real title available?)
• Inequalities for the $r$th Absolute Moment of a Sum of Random Variables, $1 \leqq r \leqq 2$
• Learning in the Presence of Malicious Errors
• Mean estimation and regression under heavy-tailed distributions: A survey
• Mean estimation with sub-Gaussian rates in polynomial time
• Resilience: a criterion for learning in the presence of arbitrary outliers
• Robust Estimation of a Location Parameter
• Robust multivariate mean estimation: the optimality of trimmed mean
• Self-Normalized Processes
• Sub-Gaussian estimators of the mean of a random matrix with heavy-tailed entries
• Sub-Gaussian mean estimators

Cited in

(10)

• Robust multivariate mean estimation: the optimality of trimmed mean
• Non-asymptotic analysis and inference for an outlyingness induced winsorized mean
• Robust distributed multicategory angle-based classification for massive data
• Robust and tuning-free sparse linear regression via square-root slope
• Accelerated convergence of error quantiles using robust randomized quasi Monte Carlo methods
• Robust and computationally efficient gradient-based estimation
• On deviation probabilities in non-parametric regression with heavy-tailed noise
• Adaptive robust confidence intervals
• Estimation beyond missing (completely) at random
• Winsorized mean estimation with heavy tails and adversarial contamination

This page was built for publication: Robust and efficient mean estimation: an approach based on the properties of self-normalized sums

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2074319)