Date:         Tue, 26 Jan 2010 18:15:32 -0800
Reply-To:     Ching <kcwong5@GMAIL.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Ching <kcwong5@GMAIL.COM>
Organization: http://groups.google.com
Subject:      variance components
Comments: To: sas-l@uga.edu
Content-Type: text/plain; charset=ISO-8859-1

Hi everyone.

Is the total subj variance equal to the sum of variance estimate of
plant and plant*subj from the Covariance Parameter Estimates from Model
(1)? If so, why is it not equal to the variance of subj from Model(2)?
Am I actually not doing it right?

Any advice is greatly appreciated. Have a lovely day.

Model (1)
proc mixed covtest cl;
class plant subj;
model log_exp= / s ddfm=satterth;
random int / subject=plant s;
random int / subject=plant*subj s;
run;

Model (2)
proc mixed covtest cl;
class subj;
model log_exp= / s ddfm=satterth;
random int / subject=subj s;
run;