Robust multivariate mean estimation: The optimality of trimmed mean

Gábor Lugosi; Shahar Mendelson

doi:10.1214/20-AOS1961

Robust multivariate mean estimation: The optimality of trimmed mean

Gábor Lugosi
, Shahar Mendelson

Research output: Contribution to journal › Article › peer-review

92 Citations (Scopus)

Abstract

We consider the problem of estimating the mean of a random vector based on i.i.d. observations and adversarial contamination. We introduce a multivariate extension of the trimmed-mean estimator and show its optimal performance under minimal conditions.

Original language	English
Pages (from-to)	393-410
Number of pages	18
Journal	Annals of Statistics
Volume	49
Issue number	1
DOIs	https://doi.org/10.1214/20-AOS1961
Publication status	Published - Feb 2021

Access to Document

10.1214/20-AOS1961

Cite this

@article{70dbef18222f46d687b3a599bdb127a5,

title = "Robust multivariate mean estimation: The optimality of trimmed mean",

abstract = "We consider the problem of estimating the mean of a random vector based on i.i.d. observations and adversarial contamination. We introduce a multivariate extension of the trimmed-mean estimator and show its optimal performance under minimal conditions.",

keywords = "Mean estimation, Robust estimation",

author = "G{\'a}bor Lugosi and Shahar Mendelson",

note = "Publisher Copyright: {\textcopyright} Institute of Mathematical Statistics, 2021",

year = "2021",

month = feb,

doi = "10.1214/20-AOS1961",

language = "English",

volume = "49",

pages = "393--410",

journal = "Annals of Statistics",

issn = "0090-5364",

publisher = "Institute of Mathematical Statistics",

number = "1",

}

TY - JOUR

T1 - Robust multivariate mean estimation

T2 - The optimality of trimmed mean

AU - Lugosi, Gábor

AU - Mendelson, Shahar

N1 - Publisher Copyright: © Institute of Mathematical Statistics, 2021

PY - 2021/2

Y1 - 2021/2

N2 - We consider the problem of estimating the mean of a random vector based on i.i.d. observations and adversarial contamination. We introduce a multivariate extension of the trimmed-mean estimator and show its optimal performance under minimal conditions.

AB - We consider the problem of estimating the mean of a random vector based on i.i.d. observations and adversarial contamination. We introduce a multivariate extension of the trimmed-mean estimator and show its optimal performance under minimal conditions.

KW - Mean estimation

KW - Robust estimation

UR - https://www.scopus.com/pages/publications/85101226574

U2 - 10.1214/20-AOS1961

DO - 10.1214/20-AOS1961

M3 - Article

SN - 0090-5364

VL - 49

SP - 393

EP - 410

JO - Annals of Statistics

JF - Annals of Statistics

IS - 1

ER -

Robust multivariate mean estimation: The optimality of trimmed mean

Abstract

Access to Document

Other files and links

Fingerprint

Cite this