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

Multiple imputation under Bayesianly smoothed pattern-mixture models for non-ignorable drop-out.

Hakan Demirtas1

  • 1Division of Epidemiology and Biostatistics, School of Public Health, University of Illinois at Chicago, Chicago, IL 60612-4336, USA. demirtas@uic.edu

Statistics in Medicine
|June 25, 2005
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

From crisis to continuity: An analysis of factors associated with COVID-19 outbreak severity in Illinois nursing homes postpandemic.

American journal of infection control·2026
Same author

Repeated home drinking water sampling to improve detection of particulate lead spikes: a simulation study.

Journal of exposure science & environmental epidemiology·2023
Same author

Conducting research on building psychosocial support for Syrian refugee families in a humanitarian emergency.

Conflict and health·2021
Same author

Optimal decision-making in oncology development programs based on probability of success for phase III utilizing phase II/III data on response and overall survival.

Pharmaceutical statistics·2020
Same author

Hip joint moments in symptomatic vs. asymptomatic people with mild radiographic hip osteoarthritis.

Journal of biomechanics·2019
Same author

Aortic valve endocarditis and cerebral mycotic aneurysm due to brucellosis.

Journal of cardiology cases·2018
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 introduces a new Bayesian approach to handle missing data in longitudinal studies, improving estimates when data drop-out is not random. The method incorporates model uncertainty for more reliable results.

Area of Science:

  • Statistics
  • Biostatistics
  • Longitudinal Data Analysis

Background:

  • Conventional pattern-mixture models are sensitive to misspecification.
  • Uncertainty in drop-out mechanisms and population models can lead to undercoverage in interval estimates.
  • Addressing non-ignorable drop-out in longitudinal studies is crucial for accurate analysis.

Purpose of the Study:

  • Develop a novel class of Bayesian random coefficient pattern-mixture models.
  • Overcome inherent inestimability problems in pattern-mixture models without hard constraints.
  • Incorporate uncertainty about model misspecification into interval estimates.

Main Methods:

  • Utilized a hierarchical Bayesian model to smooth polynomial coefficient estimates across patterns.

Related Experiment Videos

  • Employed a three-level linear mixed-effects model accommodating random variation due to drop-out groups.
  • Applied multiple imputation to integrate model uncertainty into the imputation process.

    • Main Results:

    • Demonstrated the effectiveness of the proposed Bayesian models for non-ignorable drop-out.
    • Showed that smoothing coefficients across patterns reduces sensitivity to model misspecification.
    • Validated the approach using both real and simulated longitudinal data.

    Conclusions:

    • The proposed Bayesian random coefficient pattern-mixture models effectively address non-ignorable drop-out.
    • Multiple imputation within a three-level mixed-effects model offers a robust method for handling model uncertainty.
    • This approach provides more reliable interval estimates in longitudinal studies with complex drop-out patterns.