Jove
Visualize
Contact Us

Related Experiment Videos

Tests for gaussian repeated measures with missing data in small samples.

D J Catellier1, K E Muller

  • 1Department of Biostatistics CB#7400, University of North Carolina, Chapel Hill 27599-7400, USA. diane_catellier@UNC.EDU

Statistics in Medicine
|May 3, 2000
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

Prevalence of disorders in preweaned dairy calves from 731 dairies in Germany: A cross-sectional study.

Journal of dairy science·2021
Same author

Factors associated with calf mortality and poor growth of dairy heifer calves in northeast Germany.

Preventive veterinary medicine·2020
Same author

Elevated markers of gut leakage and inflammasome activation in COVID-19 patients with cardiac involvement.

Journal of internal medicine·2020
Same author

Distinct and early increase in circulating MMP-9 in COVID-19 patients with respiratory failure.

The Journal of infection·2020
Same author

Calculating Power for the General Linear Multivariate Model With One or More Gaussian Covariates.

Communications in statistics: theory and methods·2019
Same author

On the Distribution of Summary Statistics for Missing Data.

Communications in statistics: theory and methods·2019
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
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

This study introduces new statistical tests for Gaussian repeated measures with missing data, even for small sample sizes (N=12). Adjusted tests effectively control statistical error rates, ensuring reliable results with up to 10% missing data.

Area of Science:

  • Statistics
  • Biostatistics
  • Psychometrics

Background:

  • Missing data in Gaussian repeated measures complicates statistical analysis.
  • Previous methods like EM algorithm and Rao's F approximation have limitations with small sample sizes.
  • Existing tests may not adequately control Type I error rates when data are incomplete.

Purpose of the Study:

  • To evaluate new statistical tests for Gaussian repeated measures with missing data.
  • To assess test performance with very small sample sizes (N=12, 24) and up to 10% missing data.
  • To identify optimal sample size adjustments for various multivariate tests.

Main Methods:

  • Simulated Gaussian repeated measures data with varying sample sizes and missingness.
  • Developed analogues of Pillai-Bartlett trace, Hotelling-Lawley trace, and Geisser-Greenhouse corrected tests.

Related Experiment Videos

  • Examined eleven sample size adjustment methods for error degrees of freedom.

    • Main Results:

    • Adjusted tests successfully controlled test size at or below nominal rates, even with N=12 and 10% missing data.
    • Optimal sample size adjustments varied by test statistic.
    • Mean non-missing responses per variable worked best for Geisser-Greenhouse; harmonic mean for Pillai-Bartlett; minimum non-missing pairs for Wilks' and Hotelling-Lawley.

    Conclusions:

    • Statistical tests for Gaussian repeated measures with missing data can be reliably adjusted for small sample sizes.
    • The choice of sample size adjustment is crucial and depends on the specific multivariate test used.
    • These findings offer improved methods for analyzing incomplete longitudinal data in various scientific fields.