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

Missing data perspectives of the fluvoxamine data set: a review.

G Molenberghs1, E J Goetghebeur, S R Lipsitz

  • 1Biostatistics, Limburgs Universitair Centrum, B3590 Diepenbeek, Belgium. geert.molenberghs@luc.ac.be

Statistics in Medicine
|September 4, 1999
PubMed
Summary

Fitting models to incomplete categorical data is complex, even with missing at random data. This study reviews challenges and proposes solutions using contextual information for improved accuracy.

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

Sick leave due to SARS-CoV-2 infection.

Occupational medicine (Oxford, England)·2023
Same author

Geographical variation of COVID-19 vaccination coverage, ethnic diversity and population composition in Flanders.

Vaccine: X·2022
Same author

Aligning Estimators With Estimands in Clinical Trials: Putting the ICH E9(R1) Guidelines Into Practice.

Therapeutic innovation & regulatory science·2020
Same author

Generalized pairwise comparison methods to analyze (non)prioritized composite endpoints.

Statistics in medicine·2019
Same author

Reducing surgical mortality in Scotland by use of the WHO Surgical Safety Checklist.

The British journal of surgery·2019
Same author

Recent Developments in the Prevention and Treatment of Missing Data.

Therapeutic innovation & regulatory science·2018

Area of Science:

  • Statistics
  • Biostatistics
  • Psychiatric Research

Background:

  • Modeling incomplete categorical data presents unique challenges compared to complete data.
  • Existing statistical methods may be insufficient, particularly with non-randomly missing data.
  • A multi-center trial on psychiatric symptom relief provides a relevant case study.

Purpose of the Study:

  • To review and highlight the complexities of fitting models to incomplete categorical data.
  • To demonstrate the bias in naive information matrix calculations under missing at random assumptions.
  • To illustrate issues arising from non-random missingness and explore potential solutions.

Main Methods:

  • Review of statistical literature on incomplete categorical data analysis.

Related Experiment Videos

  • Analysis of data from a multi-center psychiatric symptom relief trial.
  • Demonstration of information matrix bias under missing at random (MAR) and non-random missing (NMAR) scenarios.

    • Main Results:

    • The standard expected information matrix (naive information) is shown to be biased, even when data are missing at random (MAR).
    • Specific challenges and biases associated with non-random missingness (NMAR) assumptions are illustrated.
    • Contextual information is proposed as a method to mitigate some of these analytical problems.

    Conclusions:

    • Standard statistical approaches require careful adaptation for incomplete categorical data.
    • Bias in information matrix calculations is a significant issue even under MAR.
    • Utilizing contextual information may offer a viable strategy to address complexities in analyzing incomplete categorical data, particularly in psychiatric research.