Linearity-Inducing Priors for Poisson Parameter Estimation Under $L^{1}$ Loss

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Barnes, Leighton P., Dytso, Alex, Poor, H. Vincent
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912397857390592
author Barnes, Leighton P.
Dytso, Alex
Poor, H. Vincent
author_facet Barnes, Leighton P.
Dytso, Alex
Poor, H. Vincent
contents We study prior distributions for Poisson parameter estimation under $L^1$ loss. Specifically, we construct a new family of prior distributions whose optimal Bayesian estimators (the conditional medians) can be any prescribed increasing function that satisfies certain regularity conditions. In the case of affine estimators, this family is distinct from the usual conjugate priors, which are gamma distributions. Our prior distributions are constructed through a limiting process that matches certain moment conditions. These results provide the first explicit description of a family of distributions, beyond the conjugate priors, that satisfy the affine conditional median property; and more broadly for the Poisson noise model they can give any arbitrarily prescribed conditional median.
format Preprint
id arxiv_https___arxiv_org_abs_2505_21102
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Linearity-Inducing Priors for Poisson Parameter Estimation Under $L^{1}$ Loss
Barnes, Leighton P.
Dytso, Alex
Poor, H. Vincent
Statistics Theory
Information Theory
We study prior distributions for Poisson parameter estimation under $L^1$ loss. Specifically, we construct a new family of prior distributions whose optimal Bayesian estimators (the conditional medians) can be any prescribed increasing function that satisfies certain regularity conditions. In the case of affine estimators, this family is distinct from the usual conjugate priors, which are gamma distributions. Our prior distributions are constructed through a limiting process that matches certain moment conditions. These results provide the first explicit description of a family of distributions, beyond the conjugate priors, that satisfy the affine conditional median property; and more broadly for the Poisson noise model they can give any arbitrarily prescribed conditional median.
title Linearity-Inducing Priors for Poisson Parameter Estimation Under $L^{1}$ Loss
topic Statistics Theory
Information Theory
url https://arxiv.org/abs/2505.21102