Using ranked set sampling with binary outcomes in cluster randomized designs. Issue 3 (18th December 2019)
- Record Type:
- Journal Article
- Title:
- Using ranked set sampling with binary outcomes in cluster randomized designs. Issue 3 (18th December 2019)
- Main Title:
- Using ranked set sampling with binary outcomes in cluster randomized designs
- Authors:
- Wang, Xinlei
Wang, Mumu
Lim, Johan
Ahn, Soohyun - Abstract:
- Abstract : We study the use of ranked set sampling (RSS) with binary outcomes in cluster‐randomized designs (CRDs), where a generalized linear mixed model (GLMM) is used to model the hierarchical data structure involved. Under the GLMM‐based framework, we propose three different approaches to estimate the treatment effect, including the nonparametric (NP), maximum likelihood (ML) and pseudo likelihood (PL) estimators. We investigate their asymptotic properties and examine their finite‐sample performance via simulation. Based on these three RSS estimators, we further develop procedures for testing the existence of the treatment effect. We examine the power and size of our proposed RSS tests and compare them with existing tests based on simple random sampling (SRS). All the proposed RSS estimation and test methods are illustrated with two data examples, one for rare events and the other for non‐extreme events. Throughout our investigations, we also consider the possible effect of imperfect ranking. Among the proposed methods, we provide recommendations on whether to use RSS rather than SRS with binary outcomes in CRDs and, if yes, when to use which RSS method. The Canadian Journal of Statistics 48: 342–365; 2020 © 2019 Statistical Society of Canada Résumé : Les auteurs étudient l'usage d'échantillonnage d'ensembles ordonnés (RSS) avec une variable cible binaire pour des plans aléatoires par grappe (CRD) où un modèle linéaire mixte généralisé (GLMM) est utilisé pour saisir laAbstract : We study the use of ranked set sampling (RSS) with binary outcomes in cluster‐randomized designs (CRDs), where a generalized linear mixed model (GLMM) is used to model the hierarchical data structure involved. Under the GLMM‐based framework, we propose three different approaches to estimate the treatment effect, including the nonparametric (NP), maximum likelihood (ML) and pseudo likelihood (PL) estimators. We investigate their asymptotic properties and examine their finite‐sample performance via simulation. Based on these three RSS estimators, we further develop procedures for testing the existence of the treatment effect. We examine the power and size of our proposed RSS tests and compare them with existing tests based on simple random sampling (SRS). All the proposed RSS estimation and test methods are illustrated with two data examples, one for rare events and the other for non‐extreme events. Throughout our investigations, we also consider the possible effect of imperfect ranking. Among the proposed methods, we provide recommendations on whether to use RSS rather than SRS with binary outcomes in CRDs and, if yes, when to use which RSS method. The Canadian Journal of Statistics 48: 342–365; 2020 © 2019 Statistical Society of Canada Résumé : Les auteurs étudient l'usage d'échantillonnage d'ensembles ordonnés (RSS) avec une variable cible binaire pour des plans aléatoires par grappe (CRD) où un modèle linéaire mixte généralisé (GLMM) est utilisé pour saisir la structure hiérarchique des données. Dans le cadre des GLMM, les auteurs proposent trois approches différentes pour estimer l'effet du traitement, notamment des estimateurs non paramétriques, au maximum de vraisemblance, et basés sur la pseudo‐vraisemblance. Les auteurs en investiguent les propriétés asymptotiques et examinent leur performance sur des échantillons finis par la simulation. En se basant sur ces trois estimateurs, les auteurs développent une procédure de test pour la présence d'un effet de traitement. Ils examinent la puissance et la taille des tests proposés et les comparent à ceux basés sur un échantillonnage aléatoire simple. Tous les tests et estimateurs proposés sont illustrés à l'aide de deux sources de données, la première sur des événements rares, la seconde sur des événements non‐extrêmes. Au fur et à mesure des développements, les auteurs considèrent les conséquences d'un classement imparfait. Finalement, ils énoncent des recommandations sur les circonstances dans lesquelles chaque méthode est préférable. La revue canadienne de statistique 48: 342–365; 2020 © 2019 Société statistique du Canada … (more)
- Is Part Of:
- Canadian journal of statistics. Volume 48:Issue 3(2020)
- Journal:
- Canadian journal of statistics
- Issue:
- Volume 48:Issue 3(2020)
- Issue Display:
- Volume 48, Issue 3 (2020)
- Year:
- 2020
- Volume:
- 48
- Issue:
- 3
- Issue Sort Value:
- 2020-0048-0003-0000
- Page Start:
- 342
- Page End:
- 365
- Publication Date:
- 2019-12-18
- Subjects:
- Generalized linear mixed model -- likelihood inference -- nonparametric inference -- order statistics -- ranking error
Mathematical statistics -- Periodicals
519.5 - Journal URLs:
- http://archimede.mat.ulaval.ca/cjs/ ↗
http://onlinelibrary.wiley.com/journal/10.1002/(ISSN)1708-945X/issues ↗
http://www.jstor.org/journals/03195724.html ↗
http://onlinelibrary.wiley.com/ ↗
http://www.ingentaconnect.com/content/ssc/cjs ↗
http://www.mat.ulaval.ca/rcs/indexe.shtml ↗ - DOI:
- 10.1002/cjs.11533 ↗
- Languages:
- English
- ISSNs:
- 0319-5724
- Deposit Type:
- Legaldeposit
- View Content:
- Available online (eLD content is only available in our Reading Rooms) ↗
- Physical Locations:
- British Library DSC - 3035.760000
British Library DSC - BLDSS-3PM
British Library HMNTS - ELD Digital store - Ingest File:
- 13963.xml