To access the full text documents, please follow this link:

An empirical evaluation of small area estimators
Costa, Àlex; Satorra, Albert; Ventura, Eva
Universitat Pompeu Fabra. Departament d'Economia i Empresa
This paper investigates the comparative performance of five small areaestimators. We use Monte Carlo simulation in the context of boththeoretical and empirical populations. In addition to the direct andindirect estimators, we consider the optimal composite estimator withpopulation weights, and two composite estimators with estimatedweights: one that assumes homogeneity of within area variance andsquare bias, and another one that uses area specific estimates ofvariance and square bias. It is found that among the feasibleestimators, the best choice is the one that uses area specificestimates of variance and square bias.
Statistics, Econometrics and Quantitative Methods
regional statistics
small areas
root mean square error
indirect and composite estimators
L'accés als continguts d'aquest document queda condicionat a l'acceptació de les condicions d'ús establertes per la següent llicència Creative Commons
Working Paper

Show full item record