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

Relative risk, risk difference and rate difference models for sparse stratified data: a pseudo likelihood approach

T Stijnen1, H C Van Houwelingen

  • 1Department of Epidemiology and Biostatistics, Erasmus University Rotterdam, The Netherlands.

Statistics in Medicine
|December 30, 1993
PubMed
Summary

This study introduces a pseudo-likelihood method to address estimation inconsistencies in stratified binomial and Poisson data. This approach provides a generalized framework for Mantel-Haenszel estimators and conditional logistic regression analogues.

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

Safety of Anti-TNF Treatment in Liver Transplant Recipients: A Systematic Review and Meta-analysis.

Journal of Crohn's & colitis·2017
Same author

Suicidal ideation and subsequent completed suicide in both psychiatric and non-psychiatric populations: a meta-analysis.

Epidemiology and psychiatric sciences·2016
Same author

Meta-Regression of a Dose-Response Relationship of Methotrexate in Mono- and Combination Therapy in Disease-Modifying Antirheumatic Drug-Naive Early Rheumatoid Arthritis Patients.

Arthritis care & research·2016
Same author

Age affects joint space narrowing in patients with early active rheumatoid arthritis.

RMD open·2016
Same author

Metabolic health profile in young adults with Prader-Willi syndrome: results of a 2-year randomized, placebo-controlled, crossover GH trial.

Clinical endocrinology·2016
Same author

Tumour-free margins in vulvar squamous cell carcinoma: Does distance really matter?

European journal of cancer (Oxford, England : 1990)·2016

Area of Science:

  • Biostatistics
  • Statistical Modeling
  • Epidemiology

Background:

  • Standard maximum likelihood estimation faces challenges with large numbers of small strata due to inconsistent parameter estimates.
  • Traditional methods like conditioning on stratum events are ineffective for eliminating nuisance parameters in logistic models.
  • Existing literature lacks generalized frameworks for stratified association measures.

Purpose of the Study:

  • To develop a robust statistical method for analyzing stratified binomial and Poisson data.
  • To overcome consistency issues in parameter estimation when dealing with numerous small strata.
  • To provide a unified framework for existing estimators and develop new conditional regression analogues.

Main Methods:

  • Proposed a pseudo-likelihood method to address parameter estimation inconsistencies.

Related Experiment Videos

  • Applied the method to relative risk, risk difference (binomial), and rate difference (Poisson) models.
  • Demonstrated the method's generalizability and connection to Mantel-Haenszel estimators.

    • Main Results:

    • The pseudo maximum likelihood estimates are consistent and computable using standard statistical software.
    • The approach generalizes Mantel-Haenszel type estimators.
    • For 2x2 tables, it yields exact Mantel-Haenszel estimators for risk/rate differences and approximates the relative risk estimator.
    • Developed analogues of conditional logistic regression for relative risk, risk difference, and rate difference models.

    Conclusions:

    • The proposed pseudo-likelihood method effectively resolves consistency problems in stratified data analysis.
    • This method offers a more general framework for various association measures, extending existing estimators.
    • It provides novel conditional regression analogues, advancing statistical modeling for stratified data.