Re: Binomial

From: Arthur Gilmour <arthur.gilmour_at_AGRIC.NSW.GOV.AU>
Date: Mon, 27 Jun 2005 17:22:06 +1000

Dear Hugo,


1) Version 1.10 supports generalized models? (that section in the 1.10
version manual is missing, but is in the manual for the beta version 1.62
and older versions manuals)

** The reason that the qualifiers to fit generalized linear models were
omitted from the User Guide for Release 1.
is that the Schall method, implemented in ASReml is severly biased with
respect to variance parameters
on the underlying scale when there are many levels of the random terms,
especially for animal models.
It is not easy to quantify the bias in any particular case without doing a
simulation study.

2) How to calculate the heritability and the s.e. from a binary trait

**The usual approach is to take the residual variance as 1 (for probit,
3.29 for logistic) and then
calculate heritability in the usual way. You can do this using the .pin
file as described in the
Refernce manual.

3) Some good references regarding generalized mixed models, particularly
for genetic analysis?
** There was a flurry of activity in the late 80s. e.g. Gilmour et al
Biometricka 1885 593=599
Maybe others have a better list.

Research is continuing to try and find a better way. A common view is to
use a baysian approach
(which ASReml does not offer).

with God's love in the Name of Jesus Christ.

Arthur Gilmour PhD

PS. Check out
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; mobile: 0438 251 426
The ASReml discussion group is at
To join it,

Proposed travel:
XVI Conference of the AAABG (26 to 28 Sept, 2005, Noosa, Qld)

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 Jun 27 2005 - 17:22:06 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.)