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

Imputation and variable selection in linear regression models with missing covariates.

Xiaowei Yang1, Thomas R Belin, W John Boscardin

  • 1Department of Biostatistics, University of California, 11075 Santa Monica Boulevard, Suite 200, Los Angeles, California 90095-1772, USA. xyang@bayessoft.com

Biometrics
|July 14, 2005
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

Letter to the Editor Regarding "How Choice of Effect Measure Influences Minimally Sufficient Adjustment Sets for External Validity" by Webster-Clark and Keil (2023): The Covariate Class Taxonomy Should Be Strictly Defined With Respect To The Full Set Of Potential Outcome-Generating Covariates.

American journal of epidemiology·2026
Same author

Psychotropic Medication Trends Before and After Suicide Attempt Among Mid-to Late Life U.S. Veterans With Bipolar Affective Disorder.

The American journal of geriatric psychiatry : official journal of the American Association for Geriatric Psychiatry·2026
Same author

Disentangling Fatigue from Depression among Survivors of Severe COVID-19.

medRxiv : the preprint server for health sciences·2026
Same author

Outcomes following intradetrusor onabotulinumtoxinA injections and sacral neuromodulation in older men.

BJUI compass·2026
Same author

Modeling a mixture of ordinal and continuous repeated measures.

Journal of statistical computation and simulation·2026
Same author

A Bayesian method for analyzing combinations of continuous, ordinal, and nominal categorical data with missing values.

Journal of multivariate analysis·2026
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

This study introduces two Bayesian methods, impute-then-select (ITS) and simultaneously impute-and-select (SIAS), for handling missing covariate data in linear regression models. SIAS slightly outperforms ITS, with both methods improving upon traditional complete-case analysis.

Area of Science:

  • Statistics
  • Biostatistics
  • Data Science

Background:

  • Variable selection in linear regression models with missing covariates presents challenges for combining results from complete-data analyses.
  • Traditional methods like stepwise regression can yield different selected predictors across imputed datasets, hindering robust model selection.

Purpose of the Study:

  • To propose and evaluate two novel Bayesian strategies for addressing variable selection in the presence of missing covariates.
  • To compare the performance of these new strategies against traditional complete-case analysis with stepwise variable selection.

Main Methods:

  • Introduced two Bayesian approaches: 'impute, then select' (ITS) and 'simultaneously impute and select' (SIAS).
  • Implemented and evaluated methods using stochastic search variable selection for multivariate normal data.

Related Experiment Videos

  • Applied methods to a study on mental health services utilization among children in foster care.

    • Main Results:

    • Both ITS and SIAS demonstrated superior performance compared to complete-case analysis utilizing stepwise variable selection.
    • The SIAS strategy showed a slight performance advantage over the ITS strategy in simulation studies.
    • The proposed frameworks are adaptable for various Bayesian variable selection algorithms and data types.

    Conclusions:

    • The Bayesian strategies ITS and SIAS offer effective solutions for variable selection with missing covariates.
    • SIAS provides a marginally better approach than ITS, both outperforming conventional methods.
    • These methods provide a robust framework for statistical modeling in the presence of incomplete data.