Improved Regression-Cum-Ratio Estimators Using Information on Two Auxiliary Variables Dealing with Subsampling Technique of Non-Response

Abstract

In the present study, we have suggested some improved regression-cum-ratio-type estimators to estimate the population mean using two auxiliary variables in two different situations of non-response. The bias and mean square error are obtained under large sample approximation. After analysis of a numerical illustration, it has been found that the proposed class of estimators is more efficient than Hansen and Hurwitz (J Am Stat Assoc 41:517–529, 1946) usual unbiased estimator, conventional ratio and regression estimators, Singh and Kumar (Braz J Probab Stat 25(2):205–217, 2011) estimators, Muneer et. al. (Commun Stat Theory Methods 46(5):2181–2192, 2017) estimators, Kumar et. al. (J Stat Comput Simul 88(18):3694–3707, 2018) estimators and Akingbade and Okafer (Pak J Stat Oper Res 15(2):329–340, 2019) estimators. We also consider a simulation study under which the estimated performance of the proposed class of estimators is evaluated.