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

Non-ignorable missing covariates in generalized linear models.

S R Lipsitz1, J G Ibrahim, M H Chen

  • 1Department of Biostatistics, Harvard School of Public Health, Boston, MA 02115, USA.

Statistics in Medicine
|September 4, 1999
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 multicenter, prospective, observational study of cough variant asthma in China: clinical features, airway inflammation, and prognosis].

Zhonghua jie he he hu xi za zhi = Zhonghua jiehe he huxi zazhi = Chinese journal of tuberculosis and respiratory diseases·2025
Same author

[Advances in the application of anti-obesity medication for weight management in children and adolescents].

Zhonghua er ke za zhi = Chinese journal of pediatrics·2025
Same author

Use of respiratory signal analysis to assess severity of Brachycephalic Obstructive Airway Syndrome (BOAS) in dogs.

Veterinary journal (London, England : 1997)·2024
Same author

[Effects of high-fat and low-carbohydrate diet combined with radiotherapy on tumor microenvironment of Lewis lung cancer bearing mice].

Zhonghua zhong liu za zhi [Chinese journal of oncology]·2024
Same author

Differences of bioelectrical impedance in the development and healing phase of pressure ulcers and erythema in mouse model.

Journal of tissue viability·2024
Same author

Hyperoside attenuates carbon tetrachloride-induced hepatic fibrosis via the poly(ADP-ribose)polymerase-1-high mobility group protein 1 pathway.

European journal of pharmacology·2023
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
Same journal

A Mixture of Distributed Lag Non-Linear Models to Account for Spatially Heterogeneous Exposure-Lag-Response Associations.

Statistics in medicine·2026
Same journal

Practical Considerations for Gaussian Process Modeling for Causal Inference in Quasi-Experimental Studies With Panel Data.

Statistics in medicine·2026
Same journal

Covariate Adjustment for Wilcoxon Two Sample Statistic and Test.

Statistics in medicine·2026
See all related articles

This study introduces a new likelihood method to estimate parameters in generalized linear models with non-ignorable missing covariates. The approach uses an Expectation-Maximization algorithm for accurate statistical analysis.

Area of Science:

  • Statistics
  • Biostatistics
  • Statistical Modeling

Background:

  • Missing data in covariates can bias generalized linear model (GLM) parameter estimation.
  • Non-ignorable missingness, where the probability of missingness depends on the missing value itself, presents a significant challenge.
  • Accurate statistical inference requires methods that can appropriately handle such complex missing data mechanisms.

Purpose of the Study:

  • To develop a likelihood-based method for parameter estimation in GLMs with a single non-ignorably missing covariate.
  • To propose an Expectation-Maximization (EM) algorithm tailored for this specific missing data scenario.
  • To illustrate the methodology with a real-world example from a breast cancer clinical trial.

Main Methods:

  • A logistic model is employed to describe the probability of a covariate being missing, allowing dependence on the incomplete covariate.

Related Experiment Videos

  • The proposed method accommodates both categorical and continuous covariates.
  • For missing categorical covariates, closed-form expressions for EM algorithm steps are derived. For continuous covariates, a Monte Carlo EM algorithm utilizing the Gibbs sampler is implemented.

    • Main Results:

    • The study presents a robust statistical framework for handling non-ignorably missing covariates in GLMs.
    • The derived EM algorithm provides efficient estimation of model parameters (maximum likelihood estimates - MLEs).
    • The application to a breast cancer trial demonstrates the practical utility of the proposed method in analyzing complex clinical data.

    Conclusions:

    • The developed likelihood method and EM algorithm effectively address parameter estimation challenges posed by non-ignorably missing covariates.
    • The approach offers a valuable tool for researchers dealing with incomplete data in various scientific fields, particularly in biostatistics and clinical research.
    • Accurate handling of missing data is crucial for reliable conclusions in observational studies and clinical trials.