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

Modeling longitudinal data with nonignorable dropouts using a latent dropout class model.

Jason Roy1

  • 1Department of Biostatistics and Computational Biology, 601 Elmwood Ave., University of Rochester, Rochester, New York, USA. jason_roy@urmc.rochester.edu

Biometrics
|February 19, 2004
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

Five-Year Risk Prediction Models for Peripheral Artery Disease in Patients With Chronic Kidney Disease.

Kidney medicine·2026
Same author

Beyond Missingness: Systematizing Methods for Comprehensive Data Fitness Assessment in Clinical Research.

Journal of medical Internet research·2026
Same author

A cross-sectional assessment of the individual- and fire department-level factors affecting volunteer firefighter cardiorespiratory fitness.

Fire (Basel, Switzerland)·2025
Same author

Live microbials to boost Anti-SARS-CoV-2 immunity clinical trial (Live BASIC trial): a triple-blind randomized controlled trial.

Infection·2025
Same author

Analysis of categorical data from biological experiments with logistic regression and CMH tests.

PloS one·2025
Same author

Leveraging Machine Learning and Real-World Data to Predict Chronic Obstructive Pulmonary Disease Exacerbations.

International journal of chronic obstructive pulmonary disease·2025
Same journal

Fast penalized generalized estimating equations for large longitudinal functional datasets.

Biometrics·2026
Same journal

Causally-interpretable random-effects meta-analysis.

Biometrics·2026
Same journal

Statistical inference for mean function of partially observed functional time series.

Biometrics·2026
Same journal

Subgroup identification via Interaction Tree and Mixed Model for Repeated Measures with application to Alzheimer's disease.

Biometrics·2026
Same journal

Finite mixtures of linear quantile regressions with concomitant variables: a solution to endogeneity in longitudinal data modeling.

Biometrics·2026
Same journal

Discussion on "INTACT: a method for integration of longitudinal physical activity data from multiple sources" by Jingru Zhang, Erjia Cui, Hongzhe Li, and Haochang Shou.

Biometrics·2026
See all related articles

Latent-class models offer an alternative to pattern-mixture models for longitudinal studies with missing data. This approach improves classification accuracy and parameter identification in complex dropout scenarios.

Area of Science:

  • Statistics
  • Biostatistics
  • Longitudinal Data Analysis

Background:

  • Pattern-mixture models are used for nonignorable missing data in longitudinal studies.
  • These models assume dropout times fully determine response distributions, which can cause classification errors.
  • Challenges include weak parameter identification with sparse dropout patterns.

Purpose of the Study:

  • To propose a latent-class model as an alternative to pattern-mixture models.
  • To address limitations of pattern-mixture models in handling nonignorable missing data.
  • To provide a more robust framework for analyzing longitudinal data with dropouts.

Main Methods:

  • A latent-class model is proposed, relating dropout time to unobserved class membership.
  • A regression model for the response is specified conditional on the latent variable.

Related Experiment Videos

  • Parameter estimation uses maximum likelihood, with marginal parameters derived from conditional estimates.

    • Main Results:

    • Latent-class models offer improved classification accuracy compared to pattern-mixture models.
    • This approach enhances parameter identification, especially with numerous observation times.
    • The methodology was successfully illustrated using HIV/AIDS depression data.

    Conclusions:

    • Latent-class models provide a flexible and robust alternative for analyzing longitudinal data with nonignorable missingness.
    • This method overcomes key limitations of traditional pattern-mixture models.
    • The approach is applicable to various fields dealing with longitudinal data and dropouts.