Model-assisted estimation of small area parameters

D. Morales, M. M. Rueda García, M. D. Esteban Lefler

We introduce small area estimators of means and separable parameters. The estimators are assisted by nested error regression models and they are model-assisted counterparts of the model-based empirical best linear unbiased predictors and empirical best predictors. We study the sampling-design consistency and the asymptotic normality of the introduced estimators. Results from two simulation studies show that the new estimators present a good balance between sampling bias and mean squared error.

Palabras clave: small area estimation model-assisted estimation empirical best linear unbiased predictor empirical best predictor nested error regression model.

J. Abad González, R. Martínez

M. Boubeta Martínez, M. J. Lombardía, D. Morales

E. González-Dávila, R. Martín Rivero, D. Lorenzo Díaz, Y. González Mora, J. A. González Yanes

J. Sort Ticó, J. Jiménez Casanellas, E. Tonzan Orio