I have a dataset in which there are few continuous numerical features that distribution over them is non gaussian and this means, skewness is nonzero (positive or negative).
I read that it is good to use a log transformation to bring them to a gaussian distribution before using the features. After doing this step, I see
Nan values in the numerical data.
Is this expected?
If yes. How do you fix the
Nan values (with mean or median replacement)?
To add on - when do you remove outliers? After gaussian distribution conversion or before that??
Any immediate help on this is much appreciated. Thank you.