Robust Mean Estimation for Optimization: The Impact of Heavy Tails

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: van Parys, Bart P. G., Zwart, Bert
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866918455661297664
author van Parys, Bart P. G.
Zwart, Bert
author_facet van Parys, Bart P. G.
Zwart, Bert
contents We consider the problem of constructing a least conservative estimator of the expected value $μ$ of a non-negative heavy-tailed random variable. We require that the probability of overestimating the expected value $μ$ is kept appropriately small; a natural requirement if its subsequent use in a decision process is anticipated. In this setting, we show it is optimal to estimate $μ$ by solving a distributionally robust optimization (DRO) problem using the Kullback-Leibler (KL) divergence. We further show that the statistical properties of KL-DRO compare favorably with other estimators based on truncation, variance regularization, or Wasserstein DRO.
format Preprint
id arxiv_https___arxiv_org_abs_2503_21421
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Robust Mean Estimation for Optimization: The Impact of Heavy Tails
van Parys, Bart P. G.
Zwart, Bert
Optimization and Control
Probability
Statistics Theory
60F10, 62G35, 90C17
We consider the problem of constructing a least conservative estimator of the expected value $μ$ of a non-negative heavy-tailed random variable. We require that the probability of overestimating the expected value $μ$ is kept appropriately small; a natural requirement if its subsequent use in a decision process is anticipated. In this setting, we show it is optimal to estimate $μ$ by solving a distributionally robust optimization (DRO) problem using the Kullback-Leibler (KL) divergence. We further show that the statistical properties of KL-DRO compare favorably with other estimators based on truncation, variance regularization, or Wasserstein DRO.
title Robust Mean Estimation for Optimization: The Impact of Heavy Tails
topic Optimization and Control
Probability
Statistics Theory
60F10, 62G35, 90C17
url https://arxiv.org/abs/2503.21421