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

Inference methods for saturated models in longitudinal clinical trials with incomplete binary data.

James X Song1

  • 1Global Biometry, Bayer HealthCare, Pharma Division, West Haven, CT 06460, USA. james.song.b@bayer.com

Pharmaceutical Statistics
|November 28, 2006
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 two-stage design with two co-primary endpoints.

Contemporary clinical trials communications·2018
Same author

Phase II, Multicenter, Single-Arm, Feasibility Study of Eribulin Combined With Capecitabine for Adjuvant Treatment in Estrogen Receptor-Positive, Early-Stage Breast Cancer.

Clinical breast cancer·2015
Same author

Phase 2 study of eribulin mesylate as first-line therapy for locally recurrent or metastatic human epidermal growth factor receptor 2-negative breast cancer.

Breast cancer research and treatment·2014
Same author

A two-stage patient enrichment adaptive design in phase II oncology trials.

Contemporary clinical trials·2013
Same author

Clinical trial design for mild-to-moderate community-acquired pneumonia--an industry perspective.

Clinical infectious diseases : an official publication of the Infectious Diseases Society of America·2008
Same author

Diagnostic criteria for postconcussional syndrome after mild to moderate traumatic brain injury.

The Journal of neuropsychiatry and clinical neurosciences·2005
Same journal

Impact of Information Leakage in Platform Trials With Survival Endpoints on Type I Error Control.

Pharmaceutical statistics·2026
Same journal

Harmonic Fowlkes-Mallows Index for Medical Diagnostics Tests and Optimal Cut-Off Point Selection of Binary Diseases.

Pharmaceutical statistics·2026
Same journal

Early Phase Dose-Finding Designs for CAR-T Cell Therapies.

Pharmaceutical statistics·2026
Same journal

Optimizing Randomization Ratios in Clinical Trials With Survival Endpoints.

Pharmaceutical statistics·2026
Same journal

CUI-MET: A Clinical Utility Index Based Analysis and Decision Framework for Dose Optimization in Multiple-Dose, Multiple-Outcome Randomized Trials.

Pharmaceutical statistics·2026
Same journal

Will the Pharmaceutical Industry Need Statisticians in an AI World?

Pharmaceutical statistics·2026
See all related articles

This study addresses missing data in longitudinal binary response studies. Expectation-maximization (EM) and data augmentation (DA) algorithms provide unbiased estimates for marginal and cumulative probabilities under the missing at random (MAR) assumption.

Area of Science:

  • Biostatistics
  • Longitudinal Data Analysis
  • Statistical Modeling

Background:

  • Longitudinal studies with binary outcomes frequently require estimating positive response rates over time.
  • Missing data in such studies can lead to biased estimates using conventional methods.
  • Accurate estimation is crucial for understanding trends and making informed decisions.

Purpose of the Study:

  • To demonstrate the application of Expectation-Maximization (EM) and Data Augmentation (DA) algorithms for handling missing data in longitudinal binary response datasets.
  • To estimate marginal and cumulative probabilities accurately in the presence of incomplete data.
  • To assess the robustness of these methods under the missing at random (MAR) assumption.

Main Methods:

  • Utilized Expectation-Maximization (EM) algorithm for parameter estimation.

Related Experiment Videos

  • Employed Data Augmentation (DA) algorithm as an alternative approach.
  • Performed sensitivity analyses to evaluate method performance when the MAR assumption is violated.

    • Main Results:

    • Both EM and DA algorithms yielded unbiased estimates for marginal and cumulative probabilities.
    • The methods proved effective under the missing at random (MAR) assumption.
    • Sensitivity analyses provided insights into the behavior of estimates when MAR does not hold.

    Conclusions:

    • EM and DA algorithms are reliable methods for estimating probabilities in longitudinal binary data with missingness.
    • These statistical approaches offer valid solutions for incomplete datasets when data are MAR.
    • Further investigation is warranted for scenarios where the MAR assumption is questionable.