I would like to know how to do a two-way anova:
(1) BOTH A, B treatments are random
(2) A is fixed, B is random
Should I use PROC GLM or PROC MIXED? 
My purpose is to get a anova table for each of the above two conditions.

Can I obtain correct anova table from the following command?
(1)BOTH A, B treatments are random
~~~~~~~~
proc mixed;
classes a b;
model y = a*b;
random a b;
test h= a e=a*b;
test h=b e=a*b;
run;
proc print ;
title 'a b are both random';
run;

(2) A is fixed, B is random
proc mixed;
classes a b;
model y =a a*b;
random  b;
test h=b e=a*b; <--------------How can I get a test in treatment a?
run;
proc print ;
title 'a fixed , b random';
run;

thank you for your help!
.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
.                  http://jse.stat.ncsu.edu/                    .
=================================================================

Reply via email to