Date:         Mon, 9 May 2005 10:13:57 -0700
Reply-To:     Chunling Lu <chunling_lu@YAHOO.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
Comments:     DomainKeys? See http://antispam.yahoo.com/domainkeys
From:         Chunling Lu <chunling_lu@YAHOO.COM>
Subject:      help on PROC MIXED
In-Reply-To:  6667
Content-Type: text/plain; charset=us-ascii

Dear all,

I have three equations of random effects model:

yit1=ai1+ b1*x1 + b2*x2 + eit1

yit2 = ai2 + c1*x1 + c2*x2 + eit2

yit3 = ai3 + d1*x1 + d2*x2 + eit3,

and yi1, yi2, yi3 are correlated.

How can I use "proc mixed" to run these three equations simultaneously to get the covariance matrix of ai1, ai2 and ai3?

 Note that for each equation, each unit already has, say two observations when t=1 and t=2. The data looks like:

id     t         yit1         yit2            yit3         x1           x2

1      1         10          20             30            1             40

1      2         15          25              35           1             50

2      1          0          10               30          0             30

2      2          10         45              55           0            70

Thanks very much.

Chunling





---------------------------------
Yahoo! Mail Mobile
 Take Yahoo! Mail with you! Check email on your mobile phone.