A Comparison of Correlation Structure Selection Penalties for Generalized Estimating Equations

Philip M. Westgate; Woodrow W. Burchett

doi:10.1080/00031305.2016.1200490

A Comparison of Correlation Structure Selection Penalties for Generalized Estimating Equations

Philip M. Westgate
, Woodrow W. Burchett

Producción científica: Article › revisión exhaustiva

13 Citas (Scopus)

Resumen

Correlated data are commonly analyzed using models constructed using population-averaged generalized estimating equations (GEEs). The specification of a population-averaged GEE model includes selection of a structure describing the correlation of repeated measures. Accurate specification of this structure can improve efficiency, whereas the finite-sample estimation of nuisance correlation parameters can inflate the variances of regression parameter estimates. Therefore, correlation structure selection criteria should penalize, or account for, correlation parameter estimation. In this article, we compare recently proposed penalties in terms of their impacts on correlation structure selection and regression parameter estimation, and give practical considerations for data analysts. Supplementary materials for this article are available online.

Idioma original	English
Páginas (desde-hasta)	344-353
Número de páginas	10
Publicación	American Statistician
Volumen	71
N.º	4
DOI	https://doi.org/10.1080/00031305.2016.1200490
Estado	Published - oct 2 2017

Nota bibliográfica

Publisher Copyright:
© 2017 American Statistical Association.

Financiación

The authors gratefully acknowledge the National Center for Research Resources and the National Center for Advancing Translational Sciences, National Institutes of Health, through Grant UL1TR000117. This publication was supported by the National Center for Research Resources and the National Center for Advancing Translational Sciences, National Institutes of Health, through Grant UL1TR000117. The content is solely the responsibility of the authors and does not necessarily represent the of f icial views of the NIH. We would like to thank Dr. Richard J. Kryscio, Dr. Frederick A. Schmitt, and Dr. Erin Abner for allowing us to use the data from the PREADViSE trial, which was supported through a National Institute on Aging grant (R01 AG019241).

Financiadores	Número del financiador
National Institutes of Health (NIH)	UL1TR000117
National Institutes of Health (NIH)
National Institute on Aging	R01 AG019241
National Institute on Aging
National Center for Research Resources
National Center for Advancing Translational Sciences (NCATS)

ASJC Scopus subject areas

Statistics and Probability
General Mathematics
Statistics, Probability and Uncertainty

Acceder al documento

10.1080/00031305.2016.1200490

Otros archivos y enlaces

Citar esto

@article{c7720e9759fd40059438d763be17527e,

title = "A Comparison of Correlation Structure Selection Penalties for Generalized Estimating Equations",

abstract = "Correlated data are commonly analyzed using models constructed using population-averaged generalized estimating equations (GEEs). The specification of a population-averaged GEE model includes selection of a structure describing the correlation of repeated measures. Accurate specification of this structure can improve efficiency, whereas the finite-sample estimation of nuisance correlation parameters can inflate the variances of regression parameter estimates. Therefore, correlation structure selection criteria should penalize, or account for, correlation parameter estimation. In this article, we compare recently proposed penalties in terms of their impacts on correlation structure selection and regression parameter estimation, and give practical considerations for data analysts. Supplementary materials for this article are available online.",

keywords = "Bias-correction, Efficiency, Empirical covariance matrix, Longitudinal data",

author = "Westgate, \{Philip M.\} and Burchett, \{Woodrow W.\}",

note = "Publisher Copyright: {\textcopyright} 2017 American Statistical Association.",

year = "2017",

month = oct,

day = "2",

doi = "10.1080/00031305.2016.1200490",

language = "English",

volume = "71",

pages = "344--353",

number = "4",

}

TY - JOUR

T1 - A Comparison of Correlation Structure Selection Penalties for Generalized Estimating Equations

AU - Westgate, Philip M.

AU - Burchett, Woodrow W.

PY - 2017/10/2

Y1 - 2017/10/2

N2 - Correlated data are commonly analyzed using models constructed using population-averaged generalized estimating equations (GEEs). The specification of a population-averaged GEE model includes selection of a structure describing the correlation of repeated measures. Accurate specification of this structure can improve efficiency, whereas the finite-sample estimation of nuisance correlation parameters can inflate the variances of regression parameter estimates. Therefore, correlation structure selection criteria should penalize, or account for, correlation parameter estimation. In this article, we compare recently proposed penalties in terms of their impacts on correlation structure selection and regression parameter estimation, and give practical considerations for data analysts. Supplementary materials for this article are available online.

AB - Correlated data are commonly analyzed using models constructed using population-averaged generalized estimating equations (GEEs). The specification of a population-averaged GEE model includes selection of a structure describing the correlation of repeated measures. Accurate specification of this structure can improve efficiency, whereas the finite-sample estimation of nuisance correlation parameters can inflate the variances of regression parameter estimates. Therefore, correlation structure selection criteria should penalize, or account for, correlation parameter estimation. In this article, we compare recently proposed penalties in terms of their impacts on correlation structure selection and regression parameter estimation, and give practical considerations for data analysts. Supplementary materials for this article are available online.

KW - Bias-correction

KW - Efficiency

KW - Empirical covariance matrix

KW - Longitudinal data

UR - https://www.scopus.com/pages/publications/85041026170

UR - https://www.scopus.com/pages/publications/85041026170#tab=citedBy

U2 - 10.1080/00031305.2016.1200490

DO - 10.1080/00031305.2016.1200490

M3 - Article

AN - SCOPUS:85041026170

SN - 0003-1305

VL - 71

SP - 344

EP - 353

JO - American Statistician

JF - American Statistician

IS - 4

ER -

A Comparison of Correlation Structure Selection Penalties for Generalized Estimating Equations

Resumen

Nota bibliográfica

Financiación

ASJC Scopus subject areas

Acceder al documento

Otros archivos y enlaces

Huella

Citar esto