Variance Estimation in Ranked Set Sampling Using a Concomitant Variable

TL;DR

This paper introduces a nonparametric variance estimator for ranked set sampling and judgment post stratification that leverages a concomitant variable and kernel regression to improve variance estimation accuracy.

Contribution

It presents a novel variance estimator conditioned on observed concomitant variables using nonparametric kernel regression in RSS and JPS contexts.

Findings

Estimator effectively incorporates concomitant variables

Improves variance estimation accuracy in RSS and JPS

Utilizes nonparametric kernel regression for flexibility

Abstract

We propose a nonparametric variance estimator when ranked set sampling (RSS) and judgment post stratification (JPS) are applied by measuring a concomitant variable. Our proposed estimator is obtained by conditioning on observed concomitant values and using nonparametric kernel regression.