home *** CD-ROM | disk | FTP | other *** search
- Comments: Gated by NETNEWS@AUVM.AMERICAN.EDU
- Newsgroups: bit.listserv.sas-l
- Path: sparky!uunet!paladin.american.edu!auvm!idui1!localuser
- References: <SAS-L%92121612131499@UGA.CC.UGA.EDU>
- Message-ID: <92351.112359STATPROG@IDUI1.BITNET>
- Date: Wed, 16 Dec 1992 11:23:59 PST
- Reply-To: STATPROG@IDUI1.BITNET
- Sender: "SAS(r) Discussion" <SAS-L@UGA.BITNET>
- Comments: Warning -- original Sender: tag was NETNEWS@IDUI1
- From: STATPROG@IDUI1.BITNET
- Subject: GLS in PROC MIXED
- Lines: 49
-
- I am trying to reproduce the following IML procedure in
- PROC MIXED. It is simply a case where I have estimated
- the variance-covariance matrix and want to use it in my
- regression. I haven't had any luck with MIXED so far
- but I'm not very familiar with it.
-
- Thanks in advance
- Bill Price
- ---------------------------- cut here ---------------------
-
-
- OPTIONS PS=65 NOSOURCE NONOTES;
-
- PROC IML;
-
- X={1 1,1 2,1 3,1 4,1 5,1 6};
- Y={2,5,7,7.5,11,13};
-
- V={3 2 1 0 0 0,
- 2 2 1 1 0 0,
- 1 1 4 1 1 0,
- 0 1 1 3 2 1,
- 0 0 1 2 3 2,
- 0 0 0 1 2 2};
- PRINT V;
-
- B=INV(X`*V*X)*X`*V*Y;
-
- PRINT Y B,X;
-
- SSE=Y`*(V - V*X*INV(X`*V*X)*X`*V)*Y;
-
- SSM=INV(J(1,6,1)*V*J(6,1,1))*(Y`*V*J(6,1,1)*J(1,6,1)*V*Y);
-
- SSMOD=B`*X`*V*Y - SSM;
-
- SSTOT=Y`*V*Y - SSM;
-
- PRINT SSE SSMOD SSTOT SSM;
- * * * * * * * * * * * * * * * * * * * * * * * * *
- * (__) (__) *
- * Bill Price (oo) (00) *
- * Research Associate /-------\/ /-------( ) *
- * Ag. Statistical Programs / ║ ║║ / ║ ║║() *
- * University of Idaho * ║║----║║ * ║║----║║ *
- * ~~ ~~ ~~ ~~ *
- * statprog@IDUI1 Cow Cow Chewing *
- * statprog@idui1.csrv.uidaho.edu Marbles *
- * * * * * * * * * * * * * * * * * * * * * * * * *
-