APS
2025 APS Annual Convention · 2025
Navigating the Pitfalls of MICE Algorithms: A Comparative Evaluation of Missing Data Analytical Methods
- Jiepeng Yuan
University of Virginia - David Freire
University of Virginia - Xin Tong
University of Virginia - Noelle Hurd
University of Virginia
Abstract
While multivariate imputation by chained equations (MICE) methods are widely applied to handle missing data, different algorithms may lead to inconsistent results. We systematically evaluated the performance of MICE algorithms, compared that with FIML and listwise deletion, and concluded that imputation algorithm selection is impactful and should be carefully determined.
