The Annals of Statistics
- Ann. Statist.
- Volume 46, Number 4 (2018), 1693-1720.
Empirical Bayes estimates for a two-way cross-classified model
We develop an empirical Bayes procedure for estimating the cell means in an unbalanced, two-way additive model with fixed effects. We employ a hierarchical model, which reflects exchangeability of the effects within treatment and within block but not necessarily between them, as suggested before by Lindley and Smith [J. R. Stat. Soc., B 34 (1972) 1–41]. The hyperparameters of this hierarchical model, instead of considered fixed, are to be substituted with data-dependent values in such a way that the point risk of the empirical Bayes estimator is small. Our method chooses the hyperparameters by minimizing an unbiased risk estimate and is shown to be asymptotically optimal for the estimation problem defined above, under suitable conditions. The usual empirical Best Linear Unbiased Predictor (BLUP) is shown to be substantially different from the proposed method in the unbalanced case and, therefore, performs suboptimally. Our estimator is implemented through a computationally tractable algorithm that is scalable to work under large designs. The case of missing cell observations is treated as well.
Ann. Statist., Volume 46, Number 4 (2018), 1693-1720.
Received: May 2016
Revised: February 2017
First available in Project Euclid: 27 June 2018
Permanent link to this document
Digital Object Identifier
Mathematical Reviews number (MathSciNet)
Zentralblatt MATH identifier
Brown, Lawrence D.; Mukherjee, Gourab; Weinstein, Asaf. Empirical Bayes estimates for a two-way cross-classified model. Ann. Statist. 46 (2018), no. 4, 1693--1720. doi:10.1214/17-AOS1599. https://projecteuclid.org/euclid.aos/1530086430
- Supplement to “Empirical Bayes estimates for a two-way cross-classified model”. The supplement [Brown, Mukherjee and Weinstein (2018)] contains detailed proofs of the lemmas that were used in the Appendix for proving the results in Section 3; and derivations and further discussions on the results of Sections 2 and 4.