Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

Modelling and analysing exchangeable binary data with random cluster sizes.

Jian-Lun Xu1, Philip C Prorok

  • 1Biometry Research Group, National Cancer Institute, Executive Plaza North, Suite 3131, 6130 Executive Blvd, MSC 7354, Bethesda, MD 20892-7354, USA. jianxu@helix.nih.gov

Statistics in Medicine
|July 23, 2003
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Performance of multiple multi-cancer detection tests using a large independent reference set (Alliance A212102).

Journal of the National Cancer Institute·2026
Same author

Cohort Profile: The Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial.

International journal of epidemiology·2025
Same author

Framework to Select Multi-Cancer Detection Assays in the National Cancer Institute's Vanguard Study.

Cancer epidemiology, biomarkers & prevention : a publication of the American Association for Cancer Research, cosponsored by the American Society of Preventive Oncology·2025
Same author

Randomized Study of Metformin and Intensive Lifestyle Intervention on Cancer Incidence over 21 Years of Follow-up in the Diabetes Prevention Program.

Cancer prevention research (Philadelphia, Pa.)·2025
Same author

Design of randomized controlled trials to estimate cancer-mortality reductions from multicancer detection screening.

Journal of the National Cancer Institute·2024
Same author

Increasing power in screening trials by testing control-arm specimens: application to multicancer detection screening.

Journal of the National Cancer Institute·2024
Same journal

Interpretable Bayesian Modeling for Multireader Multicase Studies: Addressing Overdispersion and Limited Sample Size in Diagnostic Enhancement Evaluation.

Statistics in medicine·2026
Same journal

Adaptive Sequential Multiple Hypotheses Testing for Concomitant Vaccine Safety Surveillance.

Statistics in medicine·2026
Same journal

Novel Distance Regression for Repeated Outcomes With Missing Data: Applications to Longitudinal and Crossover Studies of Microbiome Beta-Diversity.

Statistics in medicine·2026
Same journal

Optimal Weighted Tests for Replication Studies and the 'Two-Trials Rule' With Multiple Hypotheses.

Statistics in medicine·2026
Same journal

Identifiable Copula-Double-Cox Models: A Fully Parametric Framework for Dependent Right-Censored Survival Data.

Statistics in medicine·2026
Same journal

Moving From Individualized Risk-Based Prevention to Benefit-Based Prevention: Estimating Individualized Life-Years Gained From Prevention Services as a Basis for Eligibility.

Statistics in medicine·2026
See all related articles

This study identifies a gap in existing methods for analyzing correlated binary data, showing that previously proposed maximum likelihood estimates (MLEs) are not accurate for unequal cluster sizes. New methods requiring numerical calculation are proposed for accurate parameter estimation in these common data types.

Area of Science:

  • Biostatistics
  • Statistical Modeling
  • Clinical Trials

Background:

  • Correlated binary data are prevalent in various studies, including clinical trials and surveys.
  • Existing analysis methods like beta-binomial models and generalized estimating equations have limitations.
  • Exchangeability of data within clusters is a common and reasonable assumption.

Purpose of the Study:

  • To identify and address a gap in the maximum likelihood estimates (MLEs) for correlated binary data with unequal cluster sizes.
  • To demonstrate that existing Bowman and George estimates are not true MLEs for the joint distribution parameters.
  • To propose accurate methods for estimating population parameters in such scenarios.

Main Methods:

  • Analysis of existing methods for correlated binary data, specifically the Bowman and George extension.

Related Experiment Videos

  • Identification of the source of the gap in their approach for unequal cluster sizes.
  • Derivation of correct MLEs, showing they generally lack closed-form solutions and require numerical computation.
  • Application of derived results and generalized estimating equations to a clinical trial dataset.

    • Main Results:

    • The Bowman and George extension for unequal cluster sizes was found to have a gap, and their estimates are not the true MLEs.
    • General MLEs for population parameters do not have closed-form solutions and necessitate numerical methods for calculation.
    • The proposed methods were applied to a clinical trial comparing cefaclor and amoxicillin for acute otitis media.

    Conclusions:

    • Accurate estimation of population parameters for correlated binary data with unequal cluster sizes requires methods beyond the Bowman and George approach.
    • Numerical computation is essential for obtaining correct MLEs in many practical situations.
    • The study provides a foundation for more robust statistical analysis in diverse research fields utilizing correlated binary data.