A Spatial Unit Level Model for Small Area Estimation

Abstract

This paper approaches the problem of small area estimation in the framework of spatially correlated data. We propose a class of estimators allowing the integration of sample information of a spatial nature. Those estimators are based on linear models with spatially correlated small area effects where the neighbourhood structure is a function of the distance between small areas. Within a Monte Carlo simulation study we analyze the merits of the proposed estimators in comparison to several traditional estimators. We conclude that the proposed estimators can compete in precision with competitive estimators, while allowing significant reductions in bias. Their merits are particularly conspicuous when analyzing their conditional properties.