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

Structured correlation in models for clustered data.

Edward C Chao1

  • 1Insightful Corporation, 1700 Westlake Avenue N. Suite 500, Seattle, WA 98109, USA. echao@insightful.com

Statistics in Medicine
|October 13, 2005
PubMed
Summary

This study introduces multiblock and multilayer correlations to model complex clustered data, improving efficiency in generalized estimating equations (GEE) for high-dimensional, multilevel datasets.

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

Collaboratively Designing an App for a More Personalized, Community-Endorsed Continuous Glucose Monitoring Onboarding Experience: An Early Study.

Journal of diabetes science and technology·2023
Same author

Zooming In, Then Out: Why We Must Apply Human-Centered Design to Transform Diabetes Technology.

Journal of diabetes science and technology·2023
Same author

A smoothed corrected score approach for proportional hazards model with misclassified discretized covariates induced by error-contaminated continuous time-dependent exposure.

Biometrics·2021
Same author

Simulation Extrapolation Method for Cox Regression Model with a Mixture of Berkson and Classical Errors in the Covariates using Calibration Data.

The international journal of biostatistics·2019
Same author

SGLT-2 Inhibitors: A New Mechanism for Glycemic Control.

Clinical diabetes : a publication of the American Diabetes Association·2015
Same author

Dapagliflozin: an evidence-based review of its potential in the treatment of type-2 diabetes.

Core evidence·2012

Area of Science:

  • Statistics
  • Biostatistics
  • Data Science

Background:

  • Correlation modeling is crucial for clustered, high-dimensional data with complex structures.
  • Existing correlation structures often fall short for unbalanced, hierarchical, and heterogeneous data.
  • Multilevel data analysis presents unique challenges due to varying cluster sizes and levels.

Purpose of the Study:

  • To propose novel correlation modeling approaches for complex clustered data.
  • To extend generalized estimating equations (GEE) for high-dimensional, multilevel data.
  • To enhance the efficiency and flexibility of correlation modeling in statistical analysis.

Main Methods:

  • Introduction of multiblock and multilayer correlation structures.
  • Application of these methods within the generalized estimating equations (GEE) framework.
  • Development of an extended estimating equation with orthogonal properties for correlation parameters.

Main Results:

  • Proposed methods demonstrate significant gains in relative efficiency compared to conventional approaches.
  • Multiblock and multilayer correlations offer superior flexibility in modeling diverse data structures.
  • The extended GEE methods efficiently handle large, complex clusters with a small number of clusters.

Conclusions:

  • Multiblock and multilayer correlations provide a flexible and efficient solution for modeling complex multilevel data.
  • These extended GEE methods are particularly valuable for high-dimensional datasets with intricate hierarchical structures.
  • The proposed approaches enhance statistical inference accuracy and computational efficiency in clustered data analysis.

Related Experiment Videos