Response probability estimation

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

This paper extends the ideas in Giommi (Proc. 45th Session of the Internat. Statistical Institute, Vol. 2 (1985) 577-578; Techniques d'enquête 13(2) (1987) 137-144) and, in Särndal and Swenson (Bull Int. Statist. Inst. 15(2) (1985) 1-16; Int. Statist. Rev. 55 (1987) 279-294). Given the parallel between a 'three-phase sampling' and a 'sampling with subsequent unit and item nonresponse', we apply results from three-phase sampling theory to nonresponse situation. To handle the practical problem of unknown distributions at the second and the third phases of selection (the response mechanisms) in the nonresponse case, we use two approaches of response probability estimation: response homogeneity groups (RHG) model (Särndal and Swenson, 1985, 1987) and the nonparametric estimation (Giommi, 1985, 1987). To motivate the three-phase selection, imputation procedures for item nonresponse are used with the RHG model for unit nonresponse. By means of a Monte Carlo study, we find that the regression-type estimators are the most precise of those studied under the two approaches of response probability estimation in terms of lower bias, mean square error and variance; variance estimator close to the true variance and achieved coverage rates closer to the nominal levels. The simulation study shows how poor the variance estimators are under the single imputation approach currently used to handle the problem of missing values.

Original languageEnglish
Pages (from-to)111-126
Number of pages16
JournalJournal of Statistical Planning and Inference
Volume59
Issue number1
DOIs
Publication statusPublished - 1 Dec 1997
Externally publishedYes

Fingerprint

Non-response
Sampling
Item Nonresponse
Variance Estimator
Imputation
Homogeneity
Mean square error
Sampling Theory
Unit
Missing Values
Nonparametric Estimation
Monte Carlo Study
Use Case
Categorical or nominal
Coverage
Regression
Simulation Study
Estimator
Unknown
Model

Cite this

@article{307b1ffecda642c6aed121320c40aa05,
title = "Response probability estimation",
abstract = "This paper extends the ideas in Giommi (Proc. 45th Session of the Internat. Statistical Institute, Vol. 2 (1985) 577-578; Techniques d'enqu{\^e}te 13(2) (1987) 137-144) and, in S{\"a}rndal and Swenson (Bull Int. Statist. Inst. 15(2) (1985) 1-16; Int. Statist. Rev. 55 (1987) 279-294). Given the parallel between a 'three-phase sampling' and a 'sampling with subsequent unit and item nonresponse', we apply results from three-phase sampling theory to nonresponse situation. To handle the practical problem of unknown distributions at the second and the third phases of selection (the response mechanisms) in the nonresponse case, we use two approaches of response probability estimation: response homogeneity groups (RHG) model (S{\"a}rndal and Swenson, 1985, 1987) and the nonparametric estimation (Giommi, 1985, 1987). To motivate the three-phase selection, imputation procedures for item nonresponse are used with the RHG model for unit nonresponse. By means of a Monte Carlo study, we find that the regression-type estimators are the most precise of those studied under the two approaches of response probability estimation in terms of lower bias, mean square error and variance; variance estimator close to the true variance and achieved coverage rates closer to the nominal levels. The simulation study shows how poor the variance estimators are under the single imputation approach currently used to handle the problem of missing values.",
keywords = "Imputation, Nonparametric estimation, Response homogeneity group, Response mechanisms, Response probabilities, Three-phase selection, Variance estimation",
author = "Th{\'e}ophile Niyonsenga",
year = "1997",
month = "12",
day = "1",
doi = "10.1016/S0378-3758(96)00094-8",
language = "English",
volume = "59",
pages = "111--126",
journal = "Journal of Statistical Planning and Inference",
issn = "0378-3758",
publisher = "Elsevier",
number = "1",

}

Response probability estimation. / Niyonsenga, Théophile.

In: Journal of Statistical Planning and Inference, Vol. 59, No. 1, 01.12.1997, p. 111-126.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Response probability estimation

AU - Niyonsenga, Théophile

PY - 1997/12/1

Y1 - 1997/12/1

N2 - This paper extends the ideas in Giommi (Proc. 45th Session of the Internat. Statistical Institute, Vol. 2 (1985) 577-578; Techniques d'enquête 13(2) (1987) 137-144) and, in Särndal and Swenson (Bull Int. Statist. Inst. 15(2) (1985) 1-16; Int. Statist. Rev. 55 (1987) 279-294). Given the parallel between a 'three-phase sampling' and a 'sampling with subsequent unit and item nonresponse', we apply results from three-phase sampling theory to nonresponse situation. To handle the practical problem of unknown distributions at the second and the third phases of selection (the response mechanisms) in the nonresponse case, we use two approaches of response probability estimation: response homogeneity groups (RHG) model (Särndal and Swenson, 1985, 1987) and the nonparametric estimation (Giommi, 1985, 1987). To motivate the three-phase selection, imputation procedures for item nonresponse are used with the RHG model for unit nonresponse. By means of a Monte Carlo study, we find that the regression-type estimators are the most precise of those studied under the two approaches of response probability estimation in terms of lower bias, mean square error and variance; variance estimator close to the true variance and achieved coverage rates closer to the nominal levels. The simulation study shows how poor the variance estimators are under the single imputation approach currently used to handle the problem of missing values.

AB - This paper extends the ideas in Giommi (Proc. 45th Session of the Internat. Statistical Institute, Vol. 2 (1985) 577-578; Techniques d'enquête 13(2) (1987) 137-144) and, in Särndal and Swenson (Bull Int. Statist. Inst. 15(2) (1985) 1-16; Int. Statist. Rev. 55 (1987) 279-294). Given the parallel between a 'three-phase sampling' and a 'sampling with subsequent unit and item nonresponse', we apply results from three-phase sampling theory to nonresponse situation. To handle the practical problem of unknown distributions at the second and the third phases of selection (the response mechanisms) in the nonresponse case, we use two approaches of response probability estimation: response homogeneity groups (RHG) model (Särndal and Swenson, 1985, 1987) and the nonparametric estimation (Giommi, 1985, 1987). To motivate the three-phase selection, imputation procedures for item nonresponse are used with the RHG model for unit nonresponse. By means of a Monte Carlo study, we find that the regression-type estimators are the most precise of those studied under the two approaches of response probability estimation in terms of lower bias, mean square error and variance; variance estimator close to the true variance and achieved coverage rates closer to the nominal levels. The simulation study shows how poor the variance estimators are under the single imputation approach currently used to handle the problem of missing values.

KW - Imputation

KW - Nonparametric estimation

KW - Response homogeneity group

KW - Response mechanisms

KW - Response probabilities

KW - Three-phase selection

KW - Variance estimation

UR - http://www.scopus.com/inward/record.url?scp=0031498447&partnerID=8YFLogxK

U2 - 10.1016/S0378-3758(96)00094-8

DO - 10.1016/S0378-3758(96)00094-8

M3 - Article

VL - 59

SP - 111

EP - 126

JO - Journal of Statistical Planning and Inference

JF - Journal of Statistical Planning and Inference

SN - 0378-3758

IS - 1

ER -