Missing data imputation using K-Nearest Neighbors Classification?
Is it currently possible to use the machine learning features in JASP for the imputation of missing data? I am thinking more specifically about the Classification methods, as my relevant variables are categorical (dichotomous). To add to the complexity, my data are longitudinal with three time points and nested within participants. I have looked into K-Nearest Neighbours Classification, but unfortunately, I do not quite understand whether what I am trying to do is possible.
Thank you for your help in advance.
Comments
New year, new possibilities! Perhaps someone in here knows the answer to my question by now?
Hi Anna,
We start by saying that I'm just another user of the program, and is not a member of the development team! But I will say that there doesn't seem to be the possibility to do any kind of imputation in Jasp at the moment. Being said this functionality has been previously requested and often supported so the development team is aware of it. This is the page you need to go to to see progress, and updates etc
[Feature Request]: Missing Data - MCAR, MAR and MNAR via MICE etc · Issue #2437 · jasp-stats/jasp-issues