Programming the statistical procedures from SAS

Proc MI, imputation for a monotone pattern

Reply
N/A
Posts: 0

Proc MI, imputation for a monotone pattern

Hi,

I understand that with a monote pattern of missing data, variables are imputed sequentially one at a time, in the order determined by VAR statement. If, for example I have "VAR Y1 Y2 Y3", then my question is whether the imputed values for Y2 are used when estimating an imputation model for Y3 (assuming Y2 is a covariate for Y3), or whether the model for Y3 is built only based on observed values of Y2.

Thanks!
Ask a Question
Discussion stats
  • 0 replies
  • 110 views
  • 0 likes
  • 1 in conversation