A Constrained Conditional Likelihood Approach for Estimating the Means of Selected Populations

TL;DR

This paper introduces a constrained conditional likelihood method for estimating means of selected populations, which adaptively shrinks or averages based on observed data separation, without priors or pre-specified selection counts.

Contribution

It proposes a novel CCMLE approach that performs simultaneous inference and adapts to data separation, avoiding prior assumptions and fixed selection parameters.

Findings

If means are close, use grand mean for estimation.

If means are far apart, shrink observed means towards each other.

Method performs well without pre-specifying the number of selected populations.

Abstract

Given p independent normal populations, we consider the problem of estimating the mean of those populations, that based on the observed data, give the strongest signals. We explicitly condition on the ranking of the sample means, and consider a constrained conditional maximum likelihood (CCMLE) approach, avoiding the use of any priors and of any sparsity requirement between the population means. Our results show that if the observed means are too close together, we should in fact use the grand mean to estimate the mean of the population with the larger sample mean. If they are separated by more than a certain threshold, we should shrink the observed means towards each other. As intuition suggests, it is only if the observed means are far apart that we should conclude that the magnitude of separation and consequent ranking are not due to chance. Unlike other methods, our approach does not need to pre-specify the number of selected populations and the proposed CCMLE is able to perform simultaneous inference. Our method, which is conceptually straightforward, can be easily adapted to incorporate other selection criteria. Selected populations, Maximum likelihood, Constrained MLE, Post-selection inference