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

Using sign score regression models to select variables in case-control studies

T W O'Gorman1

  • 1Department of Mathematical Sciences, Northern Illinois University, DeKalb 60115, USA.

Statistics in Medicine
|February 28, 1995
PubMed
Summary

The sign score regression method and Kendall

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

Using adaptive tests for the analysis of repeated measurements.

Journal of biopharmaceutical statistics·2008
Same author

Using Kendall's tau b correlations to improve variable selection methods in case-control studies.

Biometrics·1995
Same author

A comparison of two methods of estimating a common risk difference in a stratified analysis of a multicenter clinical trial.

Controlled clinical trials·1994
Same author

On the efficacy of the rank transformation in stepwise logistic and discriminant analysis.

Statistics in medicine·1993
Same author

A comparison of several tests for censored paired data.

Statistics in medicine·1992
Same author

Statistical analysis of K 2 x 2 tables: a comparative study of estimators/test statistics for association and homogeneity.

Environmental health perspectives·1990

Area of Science:

  • Biostatistics
  • Statistical Modeling

Background:

  • Variable selection is crucial in case-control studies for identifying relevant predictors.
  • Traditional methods like logistic regression may be sensitive to variable correlations.

Purpose of the Study:

  • To evaluate and compare four variable selection methods for case-control studies.
  • To identify the most effective method, particularly in the presence of variable correlations.

Main Methods:

  • Logistic regression and its rank-transformed version.
  • Kendall's tau b correlation-based method.
  • A proposed sign score regression model.
  • Simulation studies using generated case and control datasets with varying distributions and covariances.

Main Results:

  • Sign score regression and Kendall's tau b were more effective than logistic regression variants for uncorrelated or weakly correlated variables.
  • Sign score regression outperformed Kendall's tau b with normal and some log-normal variables.
  • Sign score regression demonstrated superior performance across simulations with low to moderate variable correlations.

Conclusions:

  • The sign score regression method is the most effective variable selection technique for case-control studies with correlated variables.
  • This method offers improved accuracy in identifying relevant predictors compared to logistic regression and Kendall's tau b.

Related Experiment Videos