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

Likelihood inference for exchangeable binary data with varying cluster sizes.

Catalina Stefanescu1, Bruce W Turnbull

  • 1London Business School, Regent's Park, London NW1 4SA, UK.

Biometrics
|May 24, 2003
PubMed
Summary

This study introduces methods for analyzing correlated binary data using maximum likelihood estimation. The EM algorithm is proposed for estimating parameters in familial disease and cancer prevention trials.

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

Authors' reply.

Statistics in medicine·2019
Same author

Adaptive designs from a Data Safety Monitoring Board perspective: Some controversies and some case studies.

Clinical trials (London, England)·2017
Same author

Adaptive sample size modification in clinical trials: start small then ask for more?

Statistics in medicine·2015
Same author

Group sequential enrichment design incorporating subgroup selection.

Statistics in medicine·2013
Same author

Adaptive seamless designs: selection and prospective testing of hypotheses.

Journal of biopharmaceutical statistics·2007
Same author

The analysis of stratified 2 x 2 contingency tables.

Biometrical journal. Biometrische Zeitschrift·2007

Area of Science:

  • Biostatistics
  • Statistical modeling
  • Epidemiology

Background:

  • Correlated binary data present challenges in statistical analysis.
  • Accurate estimation is crucial for understanding disease aggregation and intervention effectiveness.

Purpose of the Study:

  • To investigate maximum likelihood estimation for correlated exchangeable binary data.
  • To propose and evaluate the EM algorithm for parameter estimation.

Main Methods:

  • Utilizing saturated and unsaturated models for correlated binary data.
  • Applying the Expectation-Maximization (EM) algorithm for maximum likelihood estimation.
  • Handling independent clusters of varying sizes.

Main Results:

Related Experiment Videos

  • The EM algorithm provides a viable method for maximum likelihood estimation.
  • The proposed methodology is applicable to real-world health studies.

    • Conclusions:

    • The EM algorithm is effective for analyzing correlated binary data.
    • This approach aids in the design and analysis of familial disease studies and group randomized trials.