03-07-2016 11:21 AM
I was wondering where I could find information on how both PROC MIXED and PROC NLMIXED deal with missing data. Do they assume MAR? Do they have the same approach to dealing with missing data?
When I run mixed models in both procs, the output includes "Number of Observations Read" and "Number of Observations Used," so I want to understand what is going on under the hood with both of these procs.
Any help woud be greatly appreciated!
03-15-2016 08:35 AM
The online documentation states:
PROC MIXED handles missing level combinations of classification variables similarly to the way PROC GLM does. Both procedures delete fixed-effects parameters corresponding to missing levels in order to preserve estimability. However, PROC MIXED does not delete missing level combinations for random-effects parameters because linear combinations of the random-effects parameters are always estimable. These conventions can affect the way you specify your CONTRAST and ESTIMATE coefficients.
You might consider multiple imputation if you have missing data. SAS has the MI procedure.