Programming the statistical procedures from SAS

Missing Data Problem

Reply
New Contributor
Posts: 2

Missing Data Problem

Hi folks

Given the fact that most imputation methods assume multivariate normality, does it make sense to impute missing values on a categorical variable let's say Gender and how can that be done?

Thanks

Super User
Posts: 18,491

Re: Missing Data Problem

SAS Employee
Posts: 19

Re: Missing Data Problem

Though you could impute gender using a multiple imputation method that assumes normality, better approaches are available. The MI procedure includes fully conditional specification (FCS) multiple imputation using the FCS statement. The beauty of FCS multiple imputation is that it allows the user to employ logistic regression for imputing some variables and OLS regression for imputing others. The link below provides an example of using logistic regression to impute a nominal variable (species) and OLS regression to impute the other variables in the data set (length and width).

SAS/STAT(R) 13.1 User's Guide

I believe that the FCS statement became production in SAS/STAT 13.1. So, you'll need a recent version of SAS to perform FCS multiple imputation.

Best,

Steve

Ask a Question
Discussion stats
  • 2 replies
  • 342 views
  • 1 like
  • 3 in conversation