Predictive Estimation of Population Mean in Ranked Set Sampling

Abstract

The article presents predictive estimation of population mean of the study variable in Ranked Set Sampling (RSS). It is shown that the predictive estimators in RSS using mean per unit estimator, ratio estimator and regression estimator as predictor for non-sampled values are equivalent to the corresponding classical estimators in RSS. On the other hand, when product estimator is used as predictor, the resulting estimator differs from the classical product estimator under RSS. Expressions for the Bias and the Mean Squared Error (MSE) of the proposed estimators are obtained up to first order of approximation. A simulation study is conducted to observe the performance of estimators under predictive approach.