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

Modeling nonstationary longitudinal data.

V Núñez-Antón1, D L Zimmerman

  • 1Departamento de Econometría y Estadística, Universidad del País Vasco, Bilbao-Vizcaya, Spain.

Biometrics
|September 14, 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

Sex inequality in kidney transplantation rates.

Archives of internal medicine·2000
Same author

Periodontal regeneration of intrabony defects using resorbable membranes: determinants of the healing response. An observational clinical study.

The International journal of periodontics & restorative dentistry·2000
Same author

Geographically masking health data to preserve confidentiality.

Statistics in medicine·1999
Same author

The diving reflex: bradycardia during cold ocular irrigation.

Canadian journal of anaesthesia = Journal canadien d'anesthesie·1994
Same author

Analysis of longitudinal data with unequally spaced observations and time-dependent correlated errors.

Biometrics·1994
Same author

Evidence for peptide transport across microsomal membranes.

Proceedings of the National Academy of Sciences of the United States of America·1992

Longitudinal data analysis benefits from nonstationary models that capture changing variance and correlations over time. Antedependence models are often superior for analyzing such complex longitudinal data.

Area of Science:

  • Statistics
  • Biostatistics
  • Longitudinal Data Analysis

Background:

  • Parametric models for variance-covariance structure are crucial in longitudinal data analysis.
  • Most existing models assume second-order stationarity, limiting their applicability to nonstationary data.
  • Nonstationary data exhibit time-varying variances and correlations, necessitating flexible modeling approaches.

Purpose of the Study:

  • To review and evaluate five key nonstationary models for longitudinal data analysis.
  • To assess the strengths, limitations, and appropriate use cases for each model.
  • To demonstrate the effectiveness of nonstationary models, particularly antedependence models, in analyzing longitudinal data.

Main Methods:

  • Review of five nonstationary longitudinal data models: unstructured covariance, unstructured antedependence, structured antedependence, autoregressive integrated moving average (ARIMA) and similar, and random coefficients models.

Related Experiment Videos

  • Evaluation of model performance based on strengths, limitations, and suitability for nonstationary data.
  • Application of models to three real-world examples to illustrate fitting and comparison.

    • Main Results:

    • Antedependence models (both unstructured and structured) demonstrated superior performance in the analyzed examples.
    • Random coefficients models were found to be inferior for the studied nonstationary longitudinal data.
    • Nonstationary longitudinal data can be modeled effectively and parsimoniously using appropriate models.

    Conclusions:

    • Antedependence models offer a powerful and often superior approach for analyzing nonstationary longitudinal data.
    • Greater consideration should be given to antedependence models in the field of longitudinal data analysis.
    • The study highlights the importance of selecting models that appropriately account for nonstationarity in longitudinal data.