Using Composite Estimators to Improve both Domain and Total Area Estimation

Abstract

In this article we propose using small area estimators to improve the estimates of both the small and large area parameters. When the objective is to estimate parameters at both levels accurately, optimality is achieved by a mixed sample design of fixed and proportional allocations. In the mixed sample design, once a sample size has been determined, one fraction of it is distributed proportionally among the different small areas while the rest is evenly distributed among them. We use Monte Carlo simulations to assess the performance of the direct estimator and two composite covariant-free small area estimators, for different sample sizes and different sample distributions. Performance is measured in terms of Mean Squared Errors (MSE) of both small and large area parameters. It is found that the adoption of small area composite estimators open the possibility of 1) reducing sample size when precision is given, or 2) improving precision for a given sample size.