Dear Asremlers
Sorry but i am unclear just how to work out the estimated variance
of treatment difference in ASREML.
i.e
(1) est.var(Treat(i)-Treat(j)) where i not equal to j
As i believe the AVESED is is constant for all treatment differences. I am
interested in extrapolating each est.var for each and every treatment
difference as in (1) above and working out....
sum (i,j=1....t) [est.var(tr(i)-tr(j)]
Kind Regards
Vince
Vince Matassa
Department Of Agriculture
BIOMETRICS SECTION
University Of Queensland
Brisbane 4072
Australia