i am trying to build a logistic regression model. I have a doubt regarding using woe (weight of evidence) transformed variables. I wanted to know if it is ok to use a few woe transformed variables and a few original variables (not woe transformed) in the logistic model or is it like, if we use one woe transformed variable, we have to transform all the variables by woe for the model?
$\begingroup$
$\endgroup$
Add a comment
|
2 Answers
$\begingroup$
$\endgroup$
There is no theory/assumption behind logistic regression that insists you transform all of your variables to Weight Of Evidence.
$\begingroup$
$\endgroup$
No really, variables should be transformed on their own merit. For example WOE could only be used for categorical variables, while continuous variables could be standardised/normalised. Thus it is a choice of the modeller.
