Re: evaluating likelihood profile

From: Arthur Gilmour <arthur.gilmour_at_AGRIC.NSW.GOV.AU>
Date: Mon, 21 Mar 2005 13:02:42 +1100

Dear Bart,

Typically ASReml does a univariate analysis on the GAMMA (variance ratio)

To force it to do the analysis on the variance scale, you need to specify
the residual variance as a variance component
as follows (assuming a non spatial model).

1 1 ? # Variance header line
0 0 IDV 25 !S2==1.

where 25 is the residual variance
and S2==1 forces the analysis to be on th variance scale.

Then all variance parameters need to be on the variance (not variance
ratio) scale.

So for example
  y ~ mu sex !r sire 2 !GF
1 1
0 0 IDV 25 !S2==1

would hold the sire variacne at 2 and estimate the residual variance
conditinal on that

I trust this helps you do what you want.

Best wishes in the Name of Jesus Christ, my Saviour.
Check out

Arthur Gilmour PhD
Principal Research Scientist (Biometrics)
NSW Department of Primary Industries
Orange Agricultural Institute, Forest Rd, ORANGE, 2800, AUSTRALIA
fax: 02 6391 3899; 02 6391 3922 Australia +61
telephone work: 02 6391 3815; home: 02 6364 3288;
The ASReml discussion group
   has reactivated. To join,

Ellen is doing well after her Hip replacement on 24 Feb but
she will need crutches for a while yet.

"Ducro, Bart" <Bart.Ducro_at_WUR.NL>
Sent by: ASReml users discussion group <ASREML-L_at_AGRIC.NSW.GOV.AU>
19/03/2005 02:47 AM
Please respond to ASReml users discussion group

        Subject: evaluating likelihood profile

Dear All,
I am wondering if it is possible to fix variance estimates in univariate
models, to be able to check the likelihood (given the estimates). In
multivariate cases one could use !F in the lines for specifying the
variance structure. I thought of specifying BLUP 2, and specify the
starting value behind the starting value in the model. But what if I only
want to fix one component and not all of them?
Thanks in advance
Bart Ducro
Animal Breeding and Genetics Group
Wageningen University Research Centre
The Netherlands

This message is intended for the addressee named and may contain
confidential information. If you are not the intended recipient or
received it in error, please delete the message and notify sender. Views
expressed are those of the individual sender and are not necessarily the
views of their organisation.
Received on Sun Mar 21 2005 - 13:02:42 EST

This webpage is part of the ASReml-l discussion list archives 2004-2010. More information on ASReml can be found at the VSN website. This discussion list is now deprecated - please use the VSN forum for discussion on ASReml. (These online archives were generated using the hypermail package.)