Date: Wed, 5 Sep 2007 19:37:10 -0700
Reply-To: David L Cassell <davidlcassell@MSN.COM>
Sender: "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From: David L Cassell <davidlcassell@MSN.COM>
Subject: Re: proc mixed random effect missing value
Content-Type: text/plain; format=flowed
>I am working on a mixed model with random effects. Since there are some
>missing values in the response variable and fixed effect variables, I
>don't want to include those observations in the mixed model. Based on the
>SAS online help, "the proc mixed procedure 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".
>My question is that is there any way to also exclude the random effects of
>those observations? Or I have to program and delelte those missing obs
Why are the data points missing in the first place?
If they are not Missing At Random (MAR) or Missing Completely At Random
(MCAR), then dropping those records may be biasing your results.
So, before you start dropping every record with a missing value, please
think about the reasons why data points are missing, and consider using
multiple imputation to fill in the holes instead.
David L. Cassell
3115 NW Norwood Pl.
Corvallis OR 97330
A place for moms to take a break!