Hi,
What is the mean of PMC1?
What is rep etc? The following assumes that the only fixed effects are
distance, discharge, PMC1 and the intercept.
The line in a .pin file as
disch_13 {1:4}/4 5 9 10
Will give the lsm or adjusted means for PMC1 at 1 for dischanreg level 13.
Usually PMC1 is evaluated at it's mean in order to give an interpretation
consistent. So just need given the mean of 99.9:
disch_13 {1:4}/4 5 9 * 99.9 10
If you have more complex models I suggest that you read the SAS manual on it's
GLM procedure. It has a very nice example and explanation on doing these
calculations.
Bruce
> Dear ASREML users,
>
> I've discovered I don't know how to get predicted means & se's in ASREML
> when
> there are covariates in the model.
> Given this .sln file:
> distance 20 -0.6811 0.5922
> distance 60 0.7961 0.5098
> distance 120 1.305 0.5022
> distance 180 0.1476 0.5046
> discharge 13 0.0000 0.0000
> discharge 9 2.245 0.7518
> discharge 6 1.582 0.7398
> discharge 3 2.359 0.7351
> PMC1 1 -0.2097E-01 0.3756
> mu 1 17.97 6.119
> rep 1 -0.1847 0.3166
> rep 2 0.2763 0.3184
> rep 3 -0.9158E-01 0.3175
> Then this .pin file
>
> disch_13 {1:4}/4 5 10
> will give me a predicted mean for discharge level 13, if I ignore the
> covariate PMC1, which I should in this case, but in other cases PMC1 is
> significant.
> So how can I include PMC1 in this .pin file? I've read Chapter 5 of the
> Manual and I can't see the solution.
>
> Can anyone help me with this?
>
> Thanks
>
>
>
>
> Leigh
>
>
>
>
>
>
>
>
>
> --
> Asreml mailinglist archive: http://www.chiswick.anprod.csiro.au/lists/asreml
--
Asreml mailinglist archive: http://www.chiswick.anprod.csiro.au/lists/asreml