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

Obtaining marginal estimates from conditional categorical repeated measurements models with missing data.

J K Lindsey1

  • 1Department of Biostatistics, Limburgs Universitair Centrum, B-3590 Diepenbeek, Belgium. jlindsey@luc.ac.be

Statistics in Medicine
|March 29, 2000
PubMed
Summary

This study presents a matrix manipulation method to estimate marginal probabilities from conditional log-linear models for categorical repeated measures. This approach is crucial for longitudinal data analysis, especially when dealing with missing values and complex dependencies.

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

Models for estimating the change-point in gas exchange data.

Physiological measurement·2005
Same author

Robust estimation of the median lethal dose.

Journal of biopharmaceutical statistics·2002
Same author

Dynamic modelling of a challenge-escalation cross-over study of treatment of capsaicin-induced coughing.

Statistics in medicine·2002
Same author

Analysis of the human sex ratio by using overdispersion models.

Journal of the Royal Statistical Society. Series C, Applied statistics·2002
Same author

Some statistical issues in modelling pharmacokinetic data.

Statistics in medicine·2001
Same author

A general family of distributions for longitudinal dependence with special reference to event histories.

Statistics in medicine·2001

Area of Science:

  • Statistics
  • Biostatistics
  • Longitudinal Data Analysis

Background:

  • Log-linear models are standard for categorical repeated measurement data, offering flexibility with models like Markov chains.
  • Conditional models are common, but marginal probabilities are often needed for longitudinal studies.
  • Existing methods may not fully account for complex dependencies in repeated measures data.

Purpose of the Study:

  • To develop a matrix manipulation technique for deriving maximum likelihood estimates of marginal probabilities.
  • To extend the utility of conditional categorical repeated measures models by providing marginal probability estimates.
  • To address the need for marginal probabilities in longitudinal studies with complex data structures.

Main Methods:

  • A novel matrix manipulation method is introduced to derive maximum likelihood estimates.

Related Experiment Videos

  • The technique is applied to conditional categorical repeated measures models.
  • The method accounts for missingness dependence, serial dependence, and random effects.

    • Main Results:

    • The matrix manipulation method successfully derives maximum likelihood estimates of marginal probabilities.
    • The technique was applied to the Muscatine data set, demonstrating its practical utility.
    • The analysis successfully incorporated dependencies on previous values, serial dependence, and random effects.

    Conclusions:

    • The proposed matrix manipulation method provides a straightforward way to obtain marginal probabilities from conditional models.
    • This technique enhances the analysis of categorical repeated measures data, particularly in longitudinal studies.
    • The method offers a valuable tool for researchers analyzing complex longitudinal datasets with missing observations.