Linear Response Estimators for Singular Statistical Models
Statistics Theory
2026-05-11 v1 Machine Learning
Statistics Theory
Abstract
We define susceptibilities as a measure of the response of an observable quantity of a parameterized statistical model to a perturbation of the data for a general class of observables. We define estimators for these susceptibilities as statistics in a sequence of n data-points and prove that these estimators are consistent and asymptotically unbiased in the large n regime.
Cite
@article{arxiv.2605.07970,
title = {Linear Response Estimators for Singular Statistical Models},
author = {Chris Elliott and Daniel Murfet},
journal= {arXiv preprint arXiv:2605.07970},
year = {2026}
}
Comments
24 pages, comments welcome!