This article proposes improved estimation procedures for the population mean in the presence of non-response using auxiliary information. Based on the Hansen–Hurwitz subsampling approach, two generalized exponential-type estimators are introduced for situations where non-response occurs either only on the study variable or on both the study and auxiliary variables. The proposed estimators incorporate tuning constants and an optimization parameter to minimize the mean square error (MSE) and generate optimum versions within each class. Expressions for the bias and MSE of the estimators are derived to the first order of approximation, and the efficiency comparisons of the proposed estimators with the existing estimators are established. A comprehensive empirical evaluation demonstrates that the proposed classes consistently provide more precise estimates than the traditional estimators. The results confirm that the proposed methodology provides an efficient alternative for mean estimation under non-response settings.
Kumar et al. (Thu,) studied this question.