12-29-2016 04:06 AM
I have a variable whose 10% is missing in my training data. I want to impute this by applying predictive imputation in SAS.I have seen that in R, there is mice imputation which is very similar to what I want to do. I had a look at proc mi in SAS but it does not allow univarite imputation. Which procedure or method should I use in SAS?
12-29-2016 11:41 AM
Well, I choose all the variable to be imputed but all of them except for the one I tried to impute have no missing value, so proc mi only imputed one feature. I think problem is solved