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

Robustness of a multivariate normal approximation for imputation of incomplete binary data.

Coen A Bernaards1, Thomas R Belin, Joseph L Schafer

  • 1BioOncology Biostatistics, Genentech, Inc., South San Francisco, CA 94080, USA. bernaards.coen@gene.com

Statistics in Medicine
|July 1, 2006
PubMed
Summary

Imputing missing binary data is challenging. Adaptive rounding, a statistical approximation method, offers superior performance over simple rounding or Bernoulli draws for handling missing binary data in large datasets.

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

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 author

Impact of comorbidity on patient reported outcome measures for men with localised prostate cancer.

BMC urology·2026
Same author

Developing a propensity score protocol for evaluating the oral health consequences of methamphetamine use.

Communications in statistics. Case studies, data analysis and applications·2026
Same author

Predicting Patient-Reported Outcomes Following Treatment for Localized Prostate Cancer: Model Development and Evaluation in an International Cohort Study.

Clinical genitourinary cancer·2026

Area of Science:

  • Statistics
  • Data Science
  • Biostatistics

Background:

  • Multiple imputation simplifies missing data analysis using multivariate normal models.
  • Imputing missing binary data presents unique practical challenges not fully addressed by standard methods.

Purpose of the Study:

  • To evaluate three methods for converting multivariate normal imputed values to binary imputed values.
  • To compare the performance of simple rounding, Bernoulli draws, and adaptive rounding for binary data imputation.

Main Methods:

  • Simulation studies using a large dataset (California Healthy Kids Survey) with known population values.
  • Methods evaluated: simple rounding, Bernoulli draws, and adaptive rounding based on normal approximation to binomial distribution.
  • Bias and coverage of statistics and confidence intervals were compared against true population values.

Related Experiment Videos

Main Results:

  • All evaluated methods showed satisfactory bias and coverage properties in simulation studies.
  • Adaptive rounding demonstrated the best performance, particularly considering deficits in coverage.
  • Statistical approximation-based imputation methods are preferable to complete-case analysis or avoiding missing data.

Conclusions:

  • Adaptive rounding is a recommended approach for imputing missing binary data in applied research.
  • Statistical approximation methods provide a viable solution for complex missing data scenarios.
  • The findings support the use of advanced imputation techniques over simpler or incomplete methods.