Paper #1427

Títol:

Policy-related small-area estimation

Autor:

Nicholas Longford

Data:

Juny 2014

Resum:

A method of small-area estimation with a utility function is developed. The utility characterises a policy planned to be implemented in each area, based on the area's estimate of a key quantity. It is shown by simulations that the commonly applied composite and empirical Bayes estimators are inefficient for a wide range of asymmetric utility functions. Adaptations for limited budget to implement the policy are explored. An argument is presented for a closer integration of estimation and (regional) policy making because no single small- area estimator is suitable for a wide range of purposes.

Paraules clau:

Composition; empirical Bayes; expected loss; borrowing strength; exploiting similarity; small-area estimation; utility function.

Codis JEL:

MSC classification: 62-C12, 62-D99, 62-07.

Descarregar el paper en format PDF (228 Kb)