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

Variable selection for logistic regression using a prediction-focused information criterion.

Gerda Claeskens1, Christophe Croux, Johan Van Kerckhoven

  • 1ORSTAT and University Center for Statistics, K.U. Leuven, Naamsestraat 69, B-3000 Leuven, Belgium. gerda.claeskens@econ.kuleuven.be

Biometrics
|December 13, 2006
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

Robust and efficient estimation of nonparametric generalized linear models.

Test (Madrid, Spain)·2023
Same author

Discussion on "Model Confidence Bounds for Variable Selection" by Yang Li, Yuetian Luo, Davide Ferrari, Xiaonan Hu, and Yichen Qin.

Biometrics·2019
Same author

Robust sparse canonical correlation analysis.

BMC systems biology·2016
Same author

Mixed scale joint graphical lasso.

Biostatistics (Oxford, England)·2016
Same author

Multivariate mixtures of Erlangs for density estimation under censoring.

Lifetime data analysis·2015
Same author

Copula based flexible modeling of associations between clustered event times.

Lifetime data analysis·2015
Same journal

Fast penalized generalized estimating equations for large longitudinal functional datasets.

Biometrics·2026
Same journal

Causally-interpretable random-effects meta-analysis.

Biometrics·2026
Same journal

Statistical inference for mean function of partially observed functional time series.

Biometrics·2026
Same journal

Subgroup identification via Interaction Tree and Mixed Model for Repeated Measures with application to Alzheimer's disease.

Biometrics·2026
Same journal

Finite mixtures of linear quantile regressions with concomitant variables: a solution to endogeneity in longitudinal data modeling.

Biometrics·2026
Same journal

Discussion on "INTACT: a method for integration of longitudinal physical activity data from multiple sources" by Jingru Zhang, Erjia Cui, Hongzhe Li, and Haochang Shou.

Biometrics·2026
See all related articles

We introduce new Focused Information Criteria (FIC) for variable selection in logistic regression, offering flexible model selection based on the quantity of interest and chosen risk measures like error rate for improved prediction.

Area of Science:

  • Biostatistics
  • Statistical modeling
  • Machine learning

Background:

  • Information criteria are essential for guiding model selection in biostatistics.
  • The Focused Information Criterion (FIC) is a common tool for variable selection.
  • Existing FIC versions may not cover all desired risk measures or prediction goals.

Purpose of the Study:

  • To propose novel versions of the Focused Information Criterion (FIC) for variable selection in logistic regression.
  • To extend FIC by incorporating various risk measures beyond mean squared error, including error rate for predictive tasks.
  • To enhance model selection flexibility by considering both the quantity of interest and the chosen risk measure.

Main Methods:

  • Development of generalized FIC versions accommodating diverse risk measures, including L(p) error and error rate.

Related Experiment Videos

  • Application of standard and new FIC versions to logistic regression models.
  • Utilizing simulation studies and real-world data analysis (diabetic retinopathy) to evaluate performance.

    • Main Results:

    • The proposed FIC versions provide different variable sets depending on the estimated quantity and selected risk measure.
    • FIC incorporating error rate is particularly effective for event prediction tasks in medical applications.
    • Simulation and case study demonstrate the advantages of the generalized FIC approach.

    Conclusions:

    • New FIC versions offer a more adaptable and comprehensive approach to variable selection in logistic regression.
    • The choice of risk measure within FIC significantly impacts model selection, especially for predictive modeling.
    • The generalized FIC framework improves model selection by aligning with specific inferential or predictive goals.