Nonparametric Empirical Bayes Estimation on Heterogeneous Data
share
The simultaneous estimation of many parameters η_i, based on a corresponding set of observations x_i, for i=1,..., n, is a key research problem that has received renewed attention in the high-dimensional setting. subject to a fixed variance term σ^2. However, Many practical situations involve heterogeneous data (x_i, θ_i) where θ_i is a known nuisance parameter. Effectively pooling information across samples while correctly accounting for heterogeneity presents a significant challenge in large-scale estimation problems. We address this issue by introducing the "Nonparametric Empirical Bayes Smoothing Tweedie" (NEST) estimator, which efficiently estimates η_i and properly adjusts for heterogeneity approximating the marginal density of the data f_θ_i(x_i) and applying this density to via a generalized version of Tweedie's formula. NEST is capable of handling a wider range of settings than previously proposed heterogeneous approaches as it does not make any parametric assumptions on the prior distribution of η_i. The estimation framework is simple but general enough to accommodate any member of the exponential family of distributions. thorough study of the normal means problem subject to heterogeneous variances is presented to illustrate the proposed framework. Our theoretical results show that NEST is asymptotically optimal, while simulation studies show that it outperforms competing methods, with substantial efficiency gains in many settings. The method is demonstrated on a data set measuring the performance gap in math scores between socioeconomically advantaged and disadvantaged students in K-12 schools.
READ FULL TEXT
