Scikit-learn has function to evaluate the F-statistics for univariate feature importance feature selection. According to the web page they are calculating ANOVA F value.

If I understood correctly, the univariate feature importance checks how important a specific feature is to the target without taking into consideration the other features. If I have the features $f_1,f_2,f_3,...f_n$, the univariate feature importance will check how well $f_1$ fits into the target, $f_2$ fits with the target,... $f_n$ fits into the target?

Why do we use ANOVA F when there are only two variables that are checked against each other?

I think I need some basic background here to understand the use of ANOVA F when doing univariate feature selection.



Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.