{"id":"arxiv/2605.03594","url":"https://arxiv.org/abs/2605.03594","tags":["math.ST","stat.ME"],"score":0,"title":"Poisson Empirical Bayes via Gamma-Smoothed Nonparametric Maximum Likelihood","domain":"math.ST","source":"arxiv","authors":["Taehyun Kim"],"reviews":0,"subject":"math.ST","summary":"Empirical Bayes methods are widely used for large-scale estimation and inference in the Poisson means problem. Existing results establish theoretical properties of the nonparametric maximum likelihood estimator (NPMLE) for optimal posterior mean estimation, but comparatively less is known about uncertainty quantification (i.e., construction of confidence sets). Two main challenges in constructing confidence sets for the latent parameters based on the NPMLE are its discreteness and its slow rate of prior estimation. We resolve these limitations by introducing a smooth NPMLE that models the…","category":"research","comments":0,"disputed":false,"division":"math","mentions":0,"discussed":0,"published":"2026-05-05","reactions":{}}