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

Bayesian analysis for generalized linear models with nonignorably missing covariates.

Lan Huang1, Ming-Hui Chen, Joseph G Ibrahim

  • 1SRAB, Division of Cancer Control and Population Sciences, National Cancer Institute, 6116 Executive Boulevard, Rockville, Maryland 20852, USA.

Biometrics
|September 2, 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

A novel maternal prenatal risk index to predict mortality-weighted severe maternal morbidity at hospitalization: a retrospective cohort study.

Lancet regional health. Americas·2026
Same author

Information-Based Composite Likelihood Method for Hybrid Meta-Analysis Integrating Individual Participant Data and Aggregated Data.

Statistics in medicine·2026
Same author

Canopy2: Tumor Phylogeny Inference by Bulk DNA and Single-Cell RNA Sequencing.

Statistics in biosciences·2026
Same author

Mortality-weighted severe maternal morbidity: a novel approach to assessing maternal health outcomes.

BMC pregnancy and childbirth·2025
Same author

Pair-Feeding Study Designs Can Create Biases and Inflate Type I Error Rates: A Simulation Study.

Obesity (Silver Spring, Md.)·2025
Same author

Bayesian network meta-regression for aggregate ordinal outcomes with imprecise categories.

Journal of biopharmaceutical statistics·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

We introduce Bayesian methods for generalized linear models (GLMs) with missing covariate data. New proper priors resolve impropriety issues, enabling robust parameter estimation and improved model assessment for nonignorable missing data.

Area of Science:

  • Statistics
  • Biostatistics
  • Statistical Modeling

Background:

  • Missing data in generalized linear models (GLMs) presents estimation challenges.
  • Improper priors can lead to improper posteriors in selection models for nonignorable missing data.
  • Existing model assessment criteria may not adequately address complex missing data mechanisms.

Purpose of the Study:

  • To develop Bayesian methods for parameter estimation in GLMs with nonignorably missing covariate data.
  • To propose novel proper priors for selection model parameters to ensure proper posterior inference.
  • To extend model assessment criteria for evaluating missing data mechanisms in GLMs.

Main Methods:

  • Bayesian inference using proper priors for selection model coefficients.

Related Experiment Videos

  • Development of a novel Markov chain Monte Carlo (MCMC) sampling algorithm.
  • Extension of the weighted L measure and Deviance Information Criterion (DIC) for GLMs with missing data.

    • Main Results:

    • Demonstrated that improper uniform priors lead to improper posteriors under specific conditions.
    • Proposed robust and computationally efficient proper priors that facilitate Gibbs sampling.
    • Showcased the utility of extended weighted L measure and DIC for model assessment.
    • Validated the proposed methods through simulations and real-world datasets (melanoma and liver cancer studies).

    Conclusions:

    • The proposed Bayesian approach effectively handles nonignorably missing covariate data in GLMs.
    • Novel proper priors ensure valid posterior inference and robust estimation.
    • Extended model assessment criteria provide valuable tools for evaluating missing data mechanisms.