Classes of Population Mean Estimators using Transformed Variables in Double Sampling

Transformation techniques have been used to increase the efficiency of estimators in sample surveys. In this paper, some classes of population mean estimators using transformation on an auxiliary variable and on both the auxiliary and study variables have been proposed under double sampling. The proposed estimators’ biases and mean square errors are approximated up to the first order. A simulation study and application to a rubber production dataset have been used to illustrate the proposed estimators’ performance. The results show that they perform much better than other existing estimators under given conditions.