On enhanced exponential-cum-ratio estimators using robust measures of location

Many studies are mainly concerned with the estimation of finite population mean and the well-known preferences for it are ratio estimators. This article offers a remedy for improvement in the estimation of a study variable, based on supplementary information related to dual auxiliary variables in the context of simple random sampling without replacement scheme. We proposed two new general classes of exponential-cum-ratio estimators to estimate a finite population mean utilizing dual auxiliary variables with suitable combinations of the conventional and nonconventional measures. The expression for the mean square error (MSE) and theoretical conditions for proposed classes have been obtained for evaluation purposes. The performance of the proposed classes has been compared with existing estimators in terms of MSE. It is revealed that proposed estimators are more efficient than usual and existing estimators considered in this article. The simulation and robustness studies are also part of this article. Moreover, a variety of real data sets is also considered for an empirical study to support the theoretical results.