New Calibration Estimators of Population Variance in the Presence of Random Non-Response under Successive Sampling Scheme

Abstract

This research addresses the challenge of estimating population variance in surveys conducted over two-occasion (successive) sampling, particularly when dealing with non-response. The study introduces a traditional estimator and two new calibration-based estimators to mitigate the impact of non-response. These calibration estimators are designed to improve the accuracy and reliability of estimates derived from successive sampling surveys, where non-sampling errors can significantly distort the data and the resulting population parameters. The study provides expressions for the proposed estimators and analyzes their statistical properties. Simulation studies reveal that the calibration estimators outperform the traditional estimator in terms of bias, mean squared error, and relative absolute bias, especially when non-response rates are high.