Optimal Calibration Estimators in Survey Sampling

• Activity Number: 309
• Type: Contributed
• Date/Time: Wednesday, August 14, 2002 : 10:30 AM to 12:20 PM
• Sponsor: Section on Survey Research Methods*
• Author(s): Changbao Wu*+
• Affiliation(s): University of Waterloo
• Keywords: Asymptotic variance ; Auxiliary information ; Model calibration ; Optimal estimation ; Superpopulation

Abstract

The calibration method has gained much popularity in recent literature on survey sampling, and calibration estimators are routinely computed by many survey organizations. In this work, we present optimal calibration estimators for the finite population mean, the finite population distribution function, the population variance, variance of a linear estimator, and other quadratic finite population functions under a unified framework. The optimal pseudo empirical maximum likelihood estimators, which are asymptotically equivalent to the optimal calibration estimators, are particularly useful in estimating the distribution function, the population variance and other known non-negative quantities. The question of when and how auxiliary information can be used for both the estimation of the population mean ,using a generalized regression estimator, and the estimation of its variance through calibration is addressed clearly under this framework. Some fundamental issues in using auxiliary information from survey data are also addressed under the context of optimal estimation.