Empirical Bayes mean estimation with nonparametric errors via order statistic regression

11/14/2019
by   Nikolaos Ignatiadis, et al.
0

We study empirical Bayes estimation of the effect sizes of N units from K noisy observations on each unit. We show that it is possible to achieve near-Bayes optimal mean squared error, without any assumptions or knowledge about the effect size distribution or the noise. The noise distribution can be heteroskedastic and vary arbitrarily from unit to unit. Our proposal, which we call Aurora, leverages the replication inherent in the K observations per unit and recasts the effect size estimation problem as a general regression problem. Aurora with linear regression provably matches the performance of a wide array of estimators including the sample mean, the trimmed mean, the sample median, as well as James-Stein shrunk versions thereof. Aurora automates effect size estimation for Internet-scale datasets, as we demonstrate on Google data.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
10/13/2019

Empirical and Constrained Empirical Bayes Variance Estimation Under A One Unit Per Stratum Sample Design

A single primary sampling unit (PSU) per stratum design is a popular des...
research
03/15/2023

Mean-variance constrained priors have finite maximum Bayes risk in the normal location model

Consider a normal location model X |θ∼ N(θ, σ^2) with known σ^2. Suppose...
research
12/29/2022

Gaussian Heteroskedastic Empirical Bayes without Independence

In this note, we propose empirical Bayes methods under heteroskedastic G...
research
10/03/2019

Exploring Positive Noise in Estimation Theory

Estimation of a deterministic quantity observed in non-Gaussian additive...
research
12/20/2022

Random linear estimation with rotationally-invariant designs: Asymptotics at high temperature

We study estimation in the linear model y=Aβ^⋆+ϵ, in a Bayesian setting ...
research
08/13/2019

The bias of isotonic regression

We study the bias of the isotonic regression estimator. While there is e...
research
10/27/2020

Wide flat minima and optimal generalization in classifying high-dimensional Gaussian mixtures

We analyze the connection between minimizers with good generalizing prop...

Please sign up or login with your details

Forgot password? Click here to reset