A REVIEW ON RANKED SET SAMPLING AND ITS METHODOLOGICAL EXTENSIONS

Abstract

Ranked set sampling (RSS) is a structural sampling method used to gather observational samples, facilitating the development of efficient inferential procedures. This technique is particularly useful when directly measuring sample observations is challenging due to cost, destructiveness, or time constraints, yet ranking a set of samples is relatively easy and dependable. Recently, RSS has gained popularity as a statistical tool for estimation and inference because it encodes population structure through order statistics. Ranked units can focus on different population attributes, offering better inferences than Simple Random Sampling (SRS) of the same size. Over the past few decades, researchers have proposed and examined numerous variations of the original RSS concept, resulting in more informative samples of the underlying population. This chapter provides a review of the various adaptations of the RSS approach to statistical data analysis. These modifications or extensions of the original RSS method aim to minimize ranking errors and construct efficient parameter estimators.