So what was the model fitted? Y ~ management_group !r sire?
Do these estimates reported by ASREML contain the mean? (they do in the
First, you need to look at the estimable functions of interest and not the
solutions. For example, the differences between levels seem to be more
Second, you can not compare sire and animal models with a probit model or
really any generalized linear mixed models (excluding normality with
identity link of course). The sire model is overdispersed by definition:
3/4 of the additive genetic variance is missing since the 'residual
variance' is treated as one (probit link) in both models. This will
influence your standard errors.
On Mon, 24 Sep 2001, Arthur Gilmour wrote:
> > X-Authentication-Warning: lamb.arm.li.csiro.au: petidomo set sender to
> firstname.lastname@example.org using -f
> > Date: Mon, 24 Sep 2001 10:41:22 +1000
> > To: email@example.com
> > From: Kim Bunter <firstname.lastname@example.org>
> > Subject: Solutions for FE under link functions
> > Mime-Version: 1.0
> > Dear Arthur,
> > when 0/1 data is analysed with the probit (or logit) link function, do the
> > solutions for fixed effects have any meaning? For example, the following
> > solutions are obtained for management groups under either animal (A) or
> > sire (S) models, when assuming normality (N) or using the probit link
> > function (P).
> > AN SN SP
> > 0.33 0.33 71.61
> > 0.38 0.38 71.80
> > 0.43 0.43 71.86
> > 0.58 0.59 72.72
> I would expect Fixed effects to be similar for Sire and Animal models when
> assuming normality. I would expect the same pattern in the probit effects.
> However, probit values of 71 seem way off scale.
> 0.33 translates to probit of -0.44
> 0.38 -0.315
> 0.43 -0.176
> 0.59 0.228
> So something is wrong. It could be a problem of some fixed effects having
> all observations in the same class. How complex is the model?
> > What is the interpretation of these results, and why are SP values of such
> > a different magnitude? The differences between solutions for each level
> > also vary. Are computed F_values informative for modelling purposes when
> > using the link functions?
> The F_values are indicative but not strictly appropriate. The usual
> testing strategy is to drop a term and test the
> change in deviance (under a fixed model) as a Chi square.
> The F_value is affected by the way n which over/under dispaersion
> is handled.
> > Any comments much appreciated.
> > Cheers
> > Kim
> > Kim Bunter
> > Research Scientist
> > Animal Genetics and Breeding Unit
> > University of New England
> > Armidale, NSW, 2351
> > AUSTRALIA
> > Ph (ISD): -61-2-67733788
> > Fax (ISD): -61-2-67733266
> > email: email@example.com
> > --
> > Asreml mailinglist archive: http://www.chiswick.anprod.csiro.au/lists/asreml
> Arthur Gilmour PhD mailto:Arthur.Gilmour@agric.nsw.gov.au
> Principal Research Scientist (Biometrics) fax: <61> 2 6391 3899
> NSW Agriculture <61> 2 6391 3922
> Orange Agricultural Institute telephone work: <61> 2 6391 3815
> Forest Rd, ORANGE, 2800, AUSTRALIA home: <61> 2 6362 0046
> Cargo: <61> 2 6364 3288
> I havn't finished building my house at Cargo but am back at work.
> ASREML is still free by anonymous ftp from pub/aar on ftp.res.bbsrc.ac.uk
> or point your web browser at ftp://ftp.res.bbsrc.ac.uk/pub/aar/
> To join the asreml discussion list, send the message
> To send messages to the list, mailto:firstname.lastname@example.org.CSIRO.au
> Asreml list archive: http://www.chiswick.anprod.csiro.au/lists/asreml
> <> <> <> <> <> <> <>
> "Blessed are the peacemakers,
> for they shall be called sons of God" Jesus; Matthew 5:9
> Asreml mailinglist archive: http://www.chiswick.anprod.csiro.au/lists/asreml
Asreml mailinglist archive: http://www.chiswick.anprod.csiro.au/lists/asreml