On Study of Correlated Measurement Errors Using Ranked Set Sampling Based Logarithmic Imputation Techniques

Accepted - September 2024

Abstract

A small amount of research has been done to assess the measurement errors in the presence of missing data using a few sampling approaches, but no research has been done to assess the correlated measurement errors in the presence of missing data. The principal objective of this research is to suggest certain logarithmic imputation techniques and the resulting estimators when there are missing data in ranked set sampling, given that the data are tainted by correlated measurement errors. The developed resulting estimators' mean square error is determined to the first order approximation. A thorough simulation experiment using an artificial population is done to evaluate the effectiveness of the suggested imputation techniques and the accompanying derived estimators. The findings suggest that the proposed imputation techniques and the  resulting estimators repress the traditional imputation techniques and the resulting estimators. The proposed imputation techniques are also supplied with a real-life data application.