REGRESSION-BASED MEAN ESTIMATOR UTILIZING RANK AND EMPIRICAL DISTRIBUTION FUNCTION BOTH AS DUALS OF A SINGLE AUXILIARY VARIABLE

Abstract

This study proposes a novel regression-type estimator for estimating the finite population mean under simple random sampling by incorporating both the rank and the empirical distribution function (EDF) as duals of a single auxiliary variable. The proposed estimator utilizes the distributional properties of the auxiliary variable to enhance estimation efficiency. Theoretical properties, including the mean square error (MSE) and bias of the estimator, are derived, and its efficiency is evaluated using real-life data. The results demonstrate the practical applicability and improved accuracy of the proposed estimator in survey sampling.