Re: variance modelling
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: variance modelling



Dear Joe,

I am not sure I fully understanfd your model but gather you can
set up the design matrix with an ASREML model line like

 resp ~ mu !r sirea dama sireb damb animal
 and that you have a known covariance structure Ga [Gb, A] across
 the effects for sirea dama [sireb damb, animal]
 
Use a giv file to define Ga and Gb and a pedigree for A

The skeleton of an ASREML job is then

Example
 animal !P
 sirea 100 dama 200 sireb 40 damb 111 
 resp
A.giv   # giv file for 300 animals in sirea dama
B.giv   # giv file for 151 animals in sireb damb
Pedigree.ped
Data.dat  !VCC 1 
resp  ~ mu !r sirea dama sireb damb animal
0 0 2
sirea 1    # Define covariance structure for sirea dama
300 0 GIV1 0.1
sireb 1    # Define covariance structure for sireb damb
151 0 GIV2 0.1
+ 8  9     # !VCC equate the GIV1 and GIV2 variances.


This job uses several of the advanced features of ASREML.
You will need to read up on giv files and the !VCC qualifier
and understand that each of the GIV variance structures actually cover
two factors in the model.

Best wishes

Arthur


> X-Authentication-Warning: lamb.arm.li.csiro.au: petidomo set sender to 
asreml-owner@lamb.arm.li.csiro.au using -f
> From: Joe Daley <j.daley@cqu.edu.au>
> To: "'asreml@chiswick.anprod.CSIRO.au'" <asreml@chiswick.anprod.CSIRO.au>
> Cc: Joe Daley <j.daley@cqu.edu.au>
> Subject: variance modelling
> Date: Mon, 27 Aug 2001 14:27:54 +1000
> MIME-Version: 1.0
> 
> G'day Asreml
> 
> I am trying to model some g2 data using a mixed model
> 
> y = XB + Zu + e
> 
> where u is (2*numg1 + numg2) vector, numg1 is the number of generation
> 1 animals and numg2 is the number of generation 2 animals.  Generation 1 
> animals are comprised of sires and dams, numg1 = numsg1 + numdg1.
> 
> u can be partitioned into u = [ ua | ub | a ]T, a is a random animal effect 
> derived from the numerator relationship matrix. 
> 
> ua can be partitioned into ua = [ uas | uad ] and ub can be partitioned 
> ub = [ ubs | ubd ], here a or b distinguish between two types of effects
> and s or d distinguish between sire or dam origin of the effect.
> 
> where 
> 
> ua ~ (0, &#61555;2aGa) and ub ~ (0, &#61555;2bGb) 
> 
> this means that sire and dam a effects are coming from the same variance
> &#61555;2a with a known correlation structure Ga as sires and dams from g1
> are
> correlated with each other. Sire and dam b effects are similar.
> 
> I am fitting the asreml model
> 
> Traitname ~ mu  !r  sire.sirea  dam.dama  sire.sireb  dam.damb animal
> 
> Here sire and dam are factors with a level for each sire and dam while
> Sirea, dama sireb and damb are the respective a or b coefficients for 
> each g2 animal.
> 
> Initially I want to set &#61555;2a = &#61555;2b so that all ua and ub
> effects are coming
> From one variance, all the u's are now grouped.
> 
> How do I get asreml to model traitname this way knowing that sirea, dama,
> sireb and damb are all from the one variance and relationships exist between
> sirea and dama and sireb and damb. I want to get a blup estimate
> for each level of sire and dam for both a and b effects and overlay the 
> predetermined correlation matrices Ga and Gb.
> 
> Secondly I want to repeat this without the constraint  &#61555;2a =
> &#61555;2b , unequal
> Variances. 
> 
> Any help is appreciated.
> 
> Cheers
> Joe Daley
> PhD researcher
> Central Queensland University
> 
> 
> 
> 
> 
> --
> 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  
     subscribe
mailto:asreml-request@chiswick.anprod.CSIRO.au

To send messages to the list, mailto:asreml@chiswick.anprod.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