Synapse
⌘+K
Synapse
PulseExploreJournal ClubResearchersJournals
Instagram
HomeJournal ClubExplore
August 17, 2025Statistical Methods in Medical ResearchOpen Access

Imputation of incomplete ordinal and nominal data by predictive mean matching

View Full Paper
Ask AI
Bookmark
Share

Authors

PAPeter C. AustinSBStef van Buuren

Discussion

Loading...

Member takes

Overview

Simulations reveal predictive mean matching outperforms logistic regression for categorical variable imputation, suggesting efficiency gains.

Key Points

  • Predictive mean matching outperformed both multinomial and ordinal logistic regression for imputing categorical variables.
  • In scenarios with varied sample sizes, predictive mean matching showed faster performance, reducing processing time by a factor of 2-6.
  • Simulations compared different imputation techniques while focusing on missing non-binary categorical variables and regression models.
  • Using predictive mean matching for categorical data can lead to better quality statistical inferences and efficiency.

Cite This Study

Austin et al. (2025) studied this question.

synapsesocial.com/papers/68af453aad7bf08b1ead282ehttps://doi.org/10.1177/09622802251362642
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 4 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Logistic regression vs. predictive mean matching for imputing binary covariates2023 · 59 citations
  2. 2Additive Structure in Qualitative Data: An Alternating Least Squares Method with Optimal Scaling Features1976 · 228 citations
  3. 3Effectiveness of Public Report Cards for Improving the Quality of Cardiac Care2009 · 259 citations
  4. 4Miscellanea. Small-sample degrees of freedom with multiple imputation1999 · 840 citations