Faculty of Economics, University of Tokyo
Department of Statistics, Rutgers University
抄録
This paper studies decision theoretic properties of benchmarked estimators which are of some importance in small area estimation problems. Benchmarking is intended to improve certain aggregate properties (such as study-wide averages) when model based estimates have been applied to individual small areas. We study admissibility and minimaxity properties of such estimators by reducing the problem to one of studying these problems in a related derived problem. For certain such problems we show that unconstrained solutions in the original (unbenchmarked) problem give unconstrained Bayes, minimax or admissible estimators which automatically satisfy the benchmark constraint. We illustrate the results with several examples. Also, minimaxity of a benchmarked empirical Bayes estimator is shown in the Fay-Herriot model, a frequently used model in small area estimation.
内容記述
The title is changed as "Dominance properties of constrained Bayes and empirical Bayes estimators"; forthcoming in Bernoulli.
本文フィルはリンク先を参照のこと
雑誌名
Discussion paper series. CIRJE-F
巻
CIRJE-F-809
発行年
2011-07
書誌レコードID
AA11450569
フォーマット
application/pdf
日本十進分類法
335
出版者
日本経済国際共同センター
出版者別名
Center for International Research on the Japanese Economy
Admissibility and Minimaxity of Benchmarked Shrinkage Estimators
