Wasserstein-Cramér-Rao Theory of Unbiased Estimation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Trillos, Nicolás García, Jaffe, Adam Quinn, Sen, Bodhisattva
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914147551150080
author Trillos, Nicolás García
Jaffe, Adam Quinn
Sen, Bodhisattva
author_facet Trillos, Nicolás García
Jaffe, Adam Quinn
Sen, Bodhisattva
contents The quantity of interest in the classical Cramér-Rao theory of unbiased estimation (e.g., the Cramér-Rao lower bound, its exact attainment for exponential families, and asymptotic efficiency of maximum likelihood estimation) is the variance, which represents the instability of an estimator when its value is compared to the value for an independently-sampled data set from the same distribution. In this paper we are interested in a quantity which represents the instability of an estimator when its value is compared to the value for an infinitesimal additive perturbation of the original data set; we refer to this as the "sensitivity" of an estimator. The resulting theory of sensitivity is based on the Wasserstein geometry in the same way that the classical theory of variance is based on the Fisher-Rao (equivalently, Hellinger) geometry, and this insight allows us to determine a collection of results which are analogous to the classical case: a Wasserstein-Cramér-Rao lower bound for the sensitivity of any unbiased estimator, a characterization of models in which there exist unbiased estimators achieving the lower bound exactly, and some concrete results that show that the Wasserstein projection estimator achieves the lower bound asymptotically. We use these results to treat many statistical examples, sometimes revealing new optimality properties for existing estimators and other times revealing entirely new estimators.
format Preprint
id arxiv_https___arxiv_org_abs_2511_07414
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Wasserstein-Cramér-Rao Theory of Unbiased Estimation
Trillos, Nicolás García
Jaffe, Adam Quinn
Sen, Bodhisattva
Statistics Theory
Optimization and Control
Methodology
Machine Learning
62B11, 62F10, 62F12, 35Q49, 49Q22
The quantity of interest in the classical Cramér-Rao theory of unbiased estimation (e.g., the Cramér-Rao lower bound, its exact attainment for exponential families, and asymptotic efficiency of maximum likelihood estimation) is the variance, which represents the instability of an estimator when its value is compared to the value for an independently-sampled data set from the same distribution. In this paper we are interested in a quantity which represents the instability of an estimator when its value is compared to the value for an infinitesimal additive perturbation of the original data set; we refer to this as the "sensitivity" of an estimator. The resulting theory of sensitivity is based on the Wasserstein geometry in the same way that the classical theory of variance is based on the Fisher-Rao (equivalently, Hellinger) geometry, and this insight allows us to determine a collection of results which are analogous to the classical case: a Wasserstein-Cramér-Rao lower bound for the sensitivity of any unbiased estimator, a characterization of models in which there exist unbiased estimators achieving the lower bound exactly, and some concrete results that show that the Wasserstein projection estimator achieves the lower bound asymptotically. We use these results to treat many statistical examples, sometimes revealing new optimality properties for existing estimators and other times revealing entirely new estimators.
title Wasserstein-Cramér-Rao Theory of Unbiased Estimation
topic Statistics Theory
Optimization and Control
Methodology
Machine Learning
62B11, 62F10, 62F12, 35Q49, 49Q22
url https://arxiv.org/abs/2511.07414