Re: FW: Rejected posting to ASREML-L@AGRIC.NSW.GOV.AU

Dear Bart,

re:
I tried to run a binomial model in ASREML, but got the surprising message
that LogL is unsuitable for comparing GLMs. I know about the complexity of
analysing binomial models, but I was unaware of the fact that 'the
situation was that bad'. Is there any reference to this? And does this
mean that it is no longer possible to run binomial models in ASREML?

Response:
The LogL value reported by ASReml is for the analysis of the working
variable.
As this working variable changes between iterations and especially between
models,
the LogL values are not comparable - even to confirm convergence of a
particular model.

ASReml also calculates a 'Variance heterogenity factor' which is the
Deviance/DF
where Deviance is calculated for the underlying Maximum likelihooh model.

In a GLM, (no random effects), this deviance can be used to compare
models.
However, for GLMM (with random effects), it does not represent the total
variation
(ignoring part related to the random effects). So it also is unsuitable
as a basis
for testing in GLMM models.

Note that the GLMM procedures in ASReml are also seriously biassed when
applied
to Animal models (which is why they have not been described in the User
Guide).

Best wishes in the Name of Jesus Christ, my Saviour.
Check out my 'CargoVale' website (see below)

Arthur Gilmour PhD
Principal Research Scientist (Biometrics)
NSW Department of Primary Industries
Orange Agricultural Institute, Forest Rd, ORANGE, 2800, AUSTRALIA

mailto:Arthur.Gilmour_at_agric.nsw.gov.au
fax: 02 6391 3899; 02 6391 3922 Australia +61
telephone work: 02 6391 3815; home: 02 6364 3288;

http://www.agric.nsw.gov.au/reader/17263
http://members.dodo.net.au/~cargovale
http://www.asreml.co.uk/
The ASReml discussion group ASREML-L_at_agric.nsw.gov.au
   has reactivated. To join, mailto:arthur.gilmour_at_agric.nsw.gov.au
Cookbook: http://uncronopio.org/luis/asreml_cookbook.html

Movements: 6-11 Feb Biometrics meeting at Thredbo
  14-18 Feb Armidale Summer School
<><><><><><><><><><><><><><><><><><><><><><><><><>

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 Wed Feb 24 2005 - 11:16:38 EST