Re: bivariate animal model

Dear Roger,

I enclosed a revised job that runs.
[Renamed as bvanimal.txt because bvanimal.as was not allowed]

1)
Since the responses are on different animals, I have fixed the residual correlation to ZERO

2) Since you are using only 2 traits, you cannot use FA models.
The FA model has 4 parameters but there are only three degreesof freedom
for parameters in a 2x2 symmetric matrix.

3) The job is set up to define PATHs 2 and 3 therefore you need to specify
2 or 3 as the argum,ent.

4. 0.1 0.1 0.1
is invalid as starting values for an unstructured matrix as this is a singular matrix. If you supply
0 0 0
ASReml will supply starting values.

------------------------
Arthur Gilmour

Retired Principal Research Scientist (Biometrics)

-------------------- m2f --------------------

Sent using Mail2Forum (http://www.mail2forum.com).

Read this topic online here:
http://www.vsni.co.uk/forum/viewtopic.php?p=584#584

-------------------- m2f --------------------

Attachments:
http://www.vsni.co.uk/forum/files/bvanimal_115.txt
Received on Thu Jun 03 2009 - 04:16:36 EST