On Regression Estimators for Different Stratified Sampling Schemes

Arpita Chatterjee; Hani M. Samawi; Lili Yu; Daniel F. Linder; Jingxian Cai; Robert L. Vogel

doi:10.1080/09720510.2017.1411027

On Regression Estimators for Different Stratified Sampling Schemes

Arpita Chatterjee, Hani M. Samawi, Lili Yu, Daniel F. Linder, Jingxian Cai, Robert L. Vogel

Georgia Southern University

Research output: Contribution to journal › Article › peer-review

Abstract

Two types of stratified regression estimators for the population mean, the separate and the combined estimators, are investigated using stratified random sampling scheme (SSRS) and stratified ranked set sampling (SRSS). We derived mean and variance of the proposed estimators. In addition, we compared the performance of the regression estimators using SRSS with respect to SSRS by simulation. Our derivations and simulations revealed that our proposed estimators are unbiased and using SRSS is more efficient than using SSRS. The procedure are illustrated by using the bilirubin levels in babies in a neonatal intensive care unit data.

Original language	American English
Journal	Journal of Statistics and Management Systems
Volume	20
DOIs	https://doi.org/10.1080/09720510.2017.1411027
State	Published - Jan 1 2017

Keywords

Regression estimators
Stratified sampling schemes

DC Disciplines

Biostatistics
Environmental Public Health
Epidemiology
Public Health

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1080/09720510.2017.1411027License: CC BY

Cite this

@article{0f1f5f25585947f1a12adcd62be2ba44,

title = "On Regression Estimators for Different Stratified Sampling Schemes",

abstract = " Two types of stratified regression estimators for the population mean, the separate and the combined estimators, are investigated using stratified random sampling scheme (SSRS) and stratified ranked set sampling (SRSS). We derived mean and variance of the proposed estimators. In addition, we compared the performance of the regression estimators using SRSS with respect to SSRS by simulation. Our derivations and simulations revealed that our proposed estimators are unbiased and using SRSS is more efficient than using SSRS. The procedure are illustrated by using the bilirubin levels in babies in a neonatal intensive care unit data.",

keywords = "Regression estimators, Stratified sampling schemes",

author = "Arpita Chatterjee and Samawi, {Hani M.} and Lili Yu and Linder, {Daniel F.} and Jingxian Cai and Vogel, {Robert L.}",

year = "2017",

month = jan,

day = "1",

doi = "10.1080/09720510.2017.1411027",

language = "American English",

volume = "20",

journal = "Journal of Statistics and Management Systems",

}

TY - JOUR

T1 - On Regression Estimators for Different Stratified Sampling Schemes

AU - Chatterjee, Arpita

AU - Samawi, Hani M.

AU - Yu, Lili

AU - Linder, Daniel F.

AU - Cai, Jingxian

AU - Vogel, Robert L.

PY - 2017/1/1

Y1 - 2017/1/1

N2 - Two types of stratified regression estimators for the population mean, the separate and the combined estimators, are investigated using stratified random sampling scheme (SSRS) and stratified ranked set sampling (SRSS). We derived mean and variance of the proposed estimators. In addition, we compared the performance of the regression estimators using SRSS with respect to SSRS by simulation. Our derivations and simulations revealed that our proposed estimators are unbiased and using SRSS is more efficient than using SSRS. The procedure are illustrated by using the bilirubin levels in babies in a neonatal intensive care unit data.

AB - Two types of stratified regression estimators for the population mean, the separate and the combined estimators, are investigated using stratified random sampling scheme (SSRS) and stratified ranked set sampling (SRSS). We derived mean and variance of the proposed estimators. In addition, we compared the performance of the regression estimators using SRSS with respect to SSRS by simulation. Our derivations and simulations revealed that our proposed estimators are unbiased and using SRSS is more efficient than using SSRS. The procedure are illustrated by using the bilirubin levels in babies in a neonatal intensive care unit data.

KW - Regression estimators

KW - Stratified sampling schemes

UR - https://digitalcommons.georgiasouthern.edu/bee-facpubs/64

U2 - 10.1080/09720510.2017.1411027

DO - 10.1080/09720510.2017.1411027

M3 - Article

VL - 20

JO - Journal of Statistics and Management Systems

JF - Journal of Statistics and Management Systems

ER -

On Regression Estimators for Different Stratified Sampling Schemes

Abstract

Keywords

DC Disciplines

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this