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

Classifying individuals based on predictors of random effects. Multicenter AIDS Cohort Study

R H Lyles1, J Xu

  • 1Department of Epidemiology, Johns Hopkins School of Hygiene and Public Health, Baltimore, MD 21205, USA. rlyles@jhsph.edu

Statistics in Medicine
|February 17, 1999
PubMed
Summary

This study introduces diagnostic tools for classifying individuals based on average exposure using mixed linear models. These methods evaluate predictors to accurately categorize subjects by their true mean exposure levels.

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

Continuous treatment with nicotine increases diazepam binding inhibitor (DBI) and its mRNA in the mouse brain.

Brain research. Molecular brain research·1998
Same author

Upregulation of pleiotrophin gene expression in developing microvasculature, macrophages, and astrocytes after acute ischemic brain injury.

The Journal of neuroscience : the official journal of the Society for Neuroscience·1998
Same author

Dynamic cross-linking by alpha-actinin determines the mechanical properties of actin filament networks.

The Journal of biological chemistry·1998
Same author

Does weather confound or modify the association of particulate air pollution with mortality? An analysis of the Philadelphia data, 1973-1980.

Environmental research·1998
Same author

Screening for overlapping bacterial artificial chromosome clones by PCR analysis with an arbitrary primer.

Proceedings of the National Academy of Sciences of the United States of America·1998
Same author

Mechanical properties of actin filament networks depend on preparation, polymerization conditions, and storage of actin monomers.

Biophysical journal·1998

Area of Science:

  • Biostatistics
  • Epidemiology
  • Statistical Modeling

Background:

  • Classifying individuals by average exposure requires effective diagnostic tools.
  • Mixed linear models define true unobservable mean exposures for individuals.

Purpose of the Study:

  • To evaluate predictors of random effects as diagnostic tools for individual classification.
  • To develop and assess performance criteria for these diagnostic techniques.

Main Methods:

  • Utilized a random effects analysis of variance model for analytical and graphical assessment.
  • Compared existing and novel predictors based on specific classification criteria.
  • Incorporated estimated model parameters in a simulation study for prediction method comparison.

Related Experiment Videos

Main Results:

  • Demonstrated that predictor performance varies based on emphasized criteria.
  • Identified new alternative predictors tailored to specific performance metrics.
  • Simulation study confirmed the effectiveness of various classification criteria.

Conclusions:

  • Predictors of random effects can serve as effective diagnostic tools for classifying individuals by mean exposure.
  • The choice of criteria significantly influences the selection and performance of predictors.
  • Applied methods to classify HIV seronegative and seropositive individuals in the Multicenter AIDS Cohort Study.