| Date: | Fri, 9 Nov 2001 17:58:39 -0500 |
|---|
| Reply-To: | Michael Tseng <lzeng@ANDREW.CMU.EDU> |
|---|
| Sender: | "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU> |
|---|
| From: | Michael Tseng <lzeng@ANDREW.CMU.EDU> |
|---|
| Organization: | Carnegie Mellon, Pittsburgh, PA |
|---|
| Subject: | A question about PROC MODEL |
|---|
| Content-Type: | text/plain; charset=us-ascii; format=flowed |
|---|
Dear SAS-L,
I have a system of simultaneous equations to fit using FIML(full
information maximum likelihood) estimation method in PROC MODEL. I know
that I could use H.<variabel name> to specify the variance of the error
term for each equation, but how do I specify the covariances? I need to
know this because the error terms are correlated across equations in the
system. Thanks!
MK
|
