0
$\begingroup$

I have done a Yeo Johnson transformation by using preprocess from caret package. I have predicted the Target Variable using linear regression. Now I want to reverse the transformation for both the Target and the predicted values. I know this question has been answered before but I still couldn't solve my problem. This question was answered here-: https://stackoverflow.com/questions/44973207/revert-transformation-preprocess-caret and here-: https://rdrr.io/github/jknowles/ModelEWS/src/R/exportPrep.R I couldn't modify it for Yeo Johnson.

$\endgroup$
0
$\begingroup$

I am not sure there is an elegant caret solution, but VGAM offers what you are looking for, an inverse of the yeo-jhonson transformation : https://www.rdocumentation.org/packages/VGAM/versions/1.0-3/topics/yeo.johnson.

| improve this answer | |
$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.