Date: Tue, 28 Oct 2008 09:08:22 -0400
Reply-To: Bucher Scott <SBucher@SCHOOLS.NYC.GOV>
Sender: "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From: Bucher Scott <SBucher@SCHOOLS.NYC.GOV>
Subject: Re: Proc MI
Content-Type: text/plain; charset="us-ascii"
The example was from "Multiple Imputation for Missing Data: Concepts and
New Development" which is posted on SAS Support at:
<http://support.sas.com/rnd/app/papers/multipleimputation.pdf>. I later
realized the paper stated: "These procedures are still under development
and will be available in experimental form in Release 8.1 of the SAS
System." Apparently it was never implemented for version 9; using Proc
MI statements that actually exist has worked out much better.
From: David L Cassell [mailto:firstname.lastname@example.org]
Sent: Tuesday, October 28, 2008 1:38 AM
To: Bucher Scott
Subject: RE: Proc MI
As far as I know, MULTINORMAL is not a legal statement in SAS 9.2
What example is this, and from what book?
David L. Cassell
3115 NW Norwood Pl.
Corvallis OR 97330
> Date: Mon, 27 Oct 2008 13:44:19 -0400
> From: SBucher@SCHOOLS.NYC.GOV
> Subject: Proc MI
> To: SAS-L@LISTSERV.UGA.EDU
> I am trying to run a bit of code provided by a colleague, but I am
> receiving the following error. Having never used Proc MI and the
> is taken directly from an example in a book, I am at a loss as to what
> the problem.
> 65 Proc mi data = hs_procmi_exercise_sas seed = 37851 out = miout1;
> 66 Multinormal method = regression;
> ERROR 180-322: Statement is not valid or it is used out of proper
> 67 Var
> 68 grade8_ela_score grade8_math_score grade8_social_studies_score
> 70 ;
> 71 Run;
When your life is on the go-take your life with you. Try Windows