I'm currently working on a project in the health sciences that involves handling missing data and am seeking insights on the following:

  • Methodological Comparison: Is there any methodological paper or study that compares the random forest technique-based imputation with regression-based Multiple Imputation by Chained Equations (MICE)?
  • Performance Evaluation: In what situations does random forest-based imputation outperform regression-based MICE, and vice versa? Are there specific contexts within health science where one method is recommended over the other?
  • Health Science Applications: Based on your experience or literature, which imputation method generally performs better in health science research? Are there particular types of data or study designs where one method shows clear advantages?
  • Any references to relevant literature, personal experiences, or insights into these questions would be greatly appreciated!

    Similar questions and discussions