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

Quasi-likelihood estimation for relative risk regression models.

Rickey E Carter1, Stuart R Lipsitz, Barbara C Tilley

  • 1Department of Biometry and Epidemiology, Medical University of South Carolina, Charleston, SC 29425, USA. carterre@musc.edu

Biostatistics (Oxford, England)
|December 25, 2004
PubMed
Summary

This study introduces a novel quasi-likelihood method for relative risk regression, addressing convergence issues in clinical trials with binary outcomes. The new method provides consistent and asymptotically normal estimates for treatment effects.

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

Hybrid Quantum-Classical Model That Combines Spatial-Temporal EEG and Digitized Counterdiabatic Quantum Features for Motor Imagery Classification.

Mayo Clinic proceedings·2026
Same author

The Evolving Role of Large Language Models in Healthcare Research and Scientific Communication.

Mayo Clinic proceedings·2026
Same author

Reading, Fast and Slow: Characterizing Radiologists' Visual Search Through Abdominal CT for Detecting Hepatic Metastases.

Academic radiology·2026
Same author

Greater locus coeruleus vulnerability in atypical clinicopathologic forms of Alzheimer's disease.

Alzheimer's & dementia : the journal of the Alzheimer's Association·2026
Same author

Burden of Potentially Undiagnosed Heart Failure With Preserved Ejection Fraction in Atrial Fibrillation and Effects of Catheter Ablation: Insights From CABANA.

Journal of the American Heart Association·2026
Same author

Accuracy of Patient Self-Reported Menopause Stage: A Cross-Sectional Study.

Mayo Clinic proceedings·2026

Area of Science:

  • Biostatistics
  • Epidemiology
  • Clinical Trials

Background:

  • Relative risk is a key measure for treatment effects in randomized clinical trials with binary outcomes.
  • Log-linear models for relative risk regression can encounter convergence problems, especially with success probabilities near one.
  • Existing constrained likelihood methods also present convergence challenges.

Purpose of the Study:

  • To propose a novel statistical method to overcome convergence issues in relative risk regression for binary outcomes.
  • To provide a robust estimation technique for treatment effects in prospective studies with multiple covariates.

Main Methods:

  • A quasi-likelihood method of moments is proposed, approximating Bernoulli outcomes with a Poisson distribution.
  • A log-linear model is applied to the success probability (mean of the assumed Poisson distribution).

Related Experiment Videos

  • Poisson maximum likelihood equations are used for unconstrained estimation of regression coefficients.

    • Main Results:

    • The proposed method yields consistent and asymptotically normal estimates for regression coefficients.
    • The technique was successfully applied to a double-blinded randomized trial in primary biliary cirrhosis.

    Conclusions:

    • The quasi-likelihood method offers a viable solution for relative risk regression convergence problems.
    • This approach enhances the analysis of treatment effects in clinical trials with binary outcomes.