An Improved Compromised Imputation Method Using Ranked Set Sampling

This paper presents a modified new imputation procedure for handling missing data using Ranked Set Sampling (RSS) design. The Proposed estimator is based on Singh and Horn (2000) imputation method. Approximate expressions for the bias and mean squared error (MSE) of the proposed estimator are derived up to the first order of approximation. The optimum conditions and minimum MSE for the proposed estimators are also obtained. The efficiency of the proposed estimator is examined through numerical investigations using six real datasets and a simulation study is also performed using artificially generated datasets. The results indicate that the proposed estimators outperform all existing estimator.