Small-area estimation with spatial similarity

N.T. Longford


Small-area estimation is gaining importance in official statistics as clients of national statistical agencies demand more and more detailed information about the country, its regions (states) and smaller geographical units. Surveys can rarely be designed so that they would contain sufficient direct information about every one of the units, especially when their sizes are disparate. Small-area estimation addresses this problem by applying models that assess the extent of similarity and exploit the similarity the units.

A class of composite estimators of small-area quantities that exploit spatial (distance-related) similarity is introduced. They are based on a distribution-free model for the areas, but the estimators are aimed to have optimal design-based properties. Composition is applied also to estimating some of the global parameters on which the small-area estimators depend. The commonly adopted assumption of random effects is not necessary for exploiting the similarity of the districts (borrowing strength across the districts). The methods are applied to estimation of the mean household sizes and the proportions of single-member households in the counties (comarcas) of Catalonia.

In Computational Statistics and Data Analysis 54, 2010.