Variable reduction when there 10+ categorical variable like sex,size(small,medium,large) etc


#1

If I have more than 10+ categorical variables like sex, size(small,medium,large) , color(Red, Black, White) etc. Is there any variable reduction method like correlation, VIF which I can use to exclude the highly dependent variables?


#2

@ved - You can use
1- Chi-square test between categorical variable.
2- Find information value of variables in the data set.

Hope this helps!
Regards,
Hinduja