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

Doubly robust estimation in missing data and causal inference models.

Heejung Bang1, James M Robins

  • 1Division of Biostatistics and Epidemiology, Department of Public Health, Weill Medical College of Cornell University, New York, New York 10021, USA. heb2013@med.cornell.edu

Biometrics
|January 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

Lipoprotein(a) lipidome and chronic kidney disease: Enrichment in triacylglycerols and diacylglycerols.

Journal of clinical lipidology·2026
Same author

Global Sensitivity Analysis for Studies Extending Inferences From a Randomized Trial to a Target Population.

Statistics in medicine·2026
Same author

Authors' Response to Comments by Di Tanna et al.

Statistics in medicine·2026
Same author

Lipoprotein(a) lipidome and diet: responses to reducing saturated fat intake in African Americans in a randomized trial.

Journal of lipid research·2026
Same author

Prevalence of undiagnosed hypertension and risk assessment using a validated survey in community-based screening in Amman, Jordan.

PloS one·2026
Same author

Meta-Analysis of Cost-Effectiveness.

Statistics in medicine·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 doubly robust (DR) estimators for missing data and causal inference. These estimators offer improved consistency, providing two chances for valid statistical inference with observational data.

Area of Science:

  • Statistics
  • Biostatistics
  • Epidemiology

Background:

  • Missing data and causal inference present challenges in statistical modeling.
  • Observational data often involves uncertainty in model specification for missingness or treatment assignment.
  • Standard estimators may lack robustness when underlying models are misspecified.

Purpose of the Study:

  • To construct and evaluate doubly robust (DR) estimators for models with ignorable missing data.
  • To develop DR estimators for causal inference using observational data.
  • To enhance the reliability of statistical inference when model assumptions are uncertain.

Main Methods:

  • Development of doubly robust (DR) estimation techniques.
  • Theoretical consistency under partial model misspecification.

Related Experiment Videos

  • Simulation studies to assess finite sample performance.
  • Application to a real-world cardiovascular clinical trial.

    • Main Results:

    • DR estimators demonstrated consistency when either the missingness mechanism model or the complete data model is correctly specified.
    • DR estimators showed consistency when either the treatment assignment model or the counterfactual data model is correctly specified.
    • Simulation studies confirmed the impressive finite sample performance of DR estimators, aligning with theoretical predictions.

    Conclusions:

    • Doubly robust (DR) estimators provide a more reliable approach to statistical inference in the presence of missing data and in causal inference settings.
    • The proposed DR methods offer improved robustness against model misspecification compared to traditional estimators.
    • The successful application to a cardiovascular trial highlights the practical utility of DR estimators.