CoChat
Sign Up
Cited by 118Statistical Methods and Bayesian InferenceStatistical Methods in Clinical TrialsStatistical Methods and Inference

A comparison of two bias-corrected covariance estimators for generalized estimating equations

Lu, Bing, Preisser, John S., Qaqish, Bahjat F., Suchindran, Chirayath, Bangdiwala, Shrikant I., Wolfson, Mark·2007·Biometrics
DOI 10.1111/j.1541-0420.2007.00764.x
Open this paper in CoChat — free account

Abstract

Mancl and DeRouen (2001, Biometrics57, 126-134) and Kauermann and Carroll (2001, JASA96, 1387-1398) proposed alternative bias-corrected covariance estimators for generalized estimating equations parameter estimates of regression models for marginal means. The finite sample properties of these estimators are compared to those of the uncorrected sandwich estimator that underestimates variances in small samples. Although the formula of Mancl and DeRouen generally overestimates variances, it often leads to coverage of 95% confidence intervals near the nominal level even in some situations with as few as 10 clusters. An explanation for these seemingly contradictory results is that the tendency to undercoverage resulting from the substantial variability of sandwich estimators counteracts the impact of overcorrecting the bias. However, these positive results do not generally hold; for small cluster sizes (e.g., <10) their estimator often results in overcoverage, and the bias-corrected covariance estimator of Kauermann and Carroll may be preferred. The methods are illustrated using data from a nested cross-sectional cluster intervention trial on reducing underage drinking.

Integrity signals

Registry

Crossref · 10.1111/j.1541-0420.2007.00764.x

Cites

18 of 18 references verified0 citation errors · 0 not checkable

Statistics

Recomputing p-values needs the paper’s full text.

Claims the paper makes

This paper’s full text isn’t openly available, so its claims and self-check weren’t run. The registry and citation checks still apply.

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • GEE models adjust for dependency of repeated measures within one participant and can handle non-normally distributed data and missing data.

    “We chose GEE models because EMA data have a hierarchical structure; this type of model adjusts for dependency of repeated measures within one participant and can handle non-normally distributed data. GEE models are also suitable for dealing with missing data; it is not required to exclude participants with missing questionnaires, nor should missing questionnaires be imputed beforehand.”

    From Experienced burden of and adherence to smartphone-based ecological momentary assessment in persons with affective disorders (2020) · cites this paper as [28]

Cites

Flagged first, then by citations

Cited by

Sorted by citations

Related

Authors
Lu, B., Preisser, J. S., Qaqish, B. F., Suchindran, C., Bangdiwala, S. I., & Wolfson, M.
Affiliations
  • University of North Carolina at Chapel Hill (US)
  • Wake Forest University Health Sciences (US)
Corresponding author
Preisser, John S.
ORCID
Venue
Biometrics
Publisher
Oxford University Press (OUP)
Published
2007-03-02
Volume
63
Issue
3
Pages
935-941
ISSN
  • 0006-341X (ISSN-L)
  • 1541-0420
DOI
10.1111/j.1541-0420.2007.00764.x
OpenAlex
W2080074665
Cited by
118
References
18
Citations by year
2026: 1 · 2025: 1 · 2024: 4 · 2023: 8 · 2022: 12 · 2021: 10 · 2020: 8 · 2019: 12 · 2018: 8 · 2017: 9 · 2016: 10 · 2015: 2 · 2014: 8 · 2013: 4 · 2012: 6
FWCI
3.77
Type
journal article
Language
en
Open access
closed
Licence
onlinelibrary.wiley.com/termsAndConditions#vor (vor)
Locations
  • Biometrics (publishedVersion)
  • PubMed (publishedVersion)
  • RePEc: Research Papers in Economics (submittedVersion)
Topics
  • Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods and Bayesian Inference
  • Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods in Clinical Trials
  • Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods and Inference
Keywords
generalized estimating equations, sandwich estimator, variance estimation, confidence interval coverage, small sample properties
MeSH terms
Algorithms, Biometry, Clinical Trials as Topic, Data Interpretation, Statistical, Regression Analysis, Sensitivity and Specificity, Reproducibility of Results, Bias, Cluster Analysis

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review