Small-Disturbance Asymptotic Theory for Linear-Calibration Estimators

This article discusses the application of small-disturbance asymptotic theory for analyzing the properties of estimators in linear calibration. A family of estimators emerging from the weighted average of classical and inverse calibration estimators is proposed, and the approximations for its asymptotic bias and asymptotic mean squared error are obtained. Based on these expressions, the properties of this family—as well as an asymptotically unbiased calibration estimator, the classical calibration estimator, and the inverse calibration estimator—are analyzed and conditions for preferring one to the other are obtained. We present an example for illustration.