Re: Negative Var
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Negative Var



Dear Hugo,
Sorry I was not able to reply earlier.

> 
> Hello Asremlers:
> 
> I am using this model:
> 200-day weight
>  ani    !P 
>  mat    !P 
>  pe     6242
>  sex    2 !A
>  sy     21 !A 
>  AdG
>  AdPH
>  AdRA
>  AdS
>  AmS
>  Dd
>  w200        
> PE2_2.prn !MAKE
> W2_1.prn 
> w200 ~ mu sex sy AdG AdPH AdRA AdS AmS Dd !r ani AmS/mat pe .4097 
> 1 1 1
> 0
> ani 2 
> 2 0 US .4081 .0 .287
> ani
> 
> And I got a negative Var estimate for the maternal effect:
> 
>  Source       Model  terms     Gamma     Component    Compnt/StndErr
>   pe             6242   1688  0.101417       78.3356          2.79
>   Variance       3224   3196   1.00000       772.413         19.69
>   ani         US=UnStr     1  0.307791       237.741          5.11
>   ani         US=UnStr     1  0.307736E-01   23.7699          1.06
>   ani         US=UnStr     2 -0.136305      -105.284        -21.63
> 
> The same happens with a DIAG structure of G,
>

Dear Hugo,
The problem is probably related to whatever  AmS/mat is forming in the design
matrix.  It is not obvious to me what you intend because I do not know what
AmS contains but AmS/mat  will be a modified form of the design
matrix generated by 'mat' with rows blanked out wherever the covariate
AmS is not 1 [1.00 up to 1.999999 actually].  Is that really what you wanted?

Arthur

<><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>
Arthur Gilmour PhD                    email: Arthur.Gilmour@agric.nsw.gov.au
Senior 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

ASREML is currently free by anonymous ftp from pub/aar on ftp.res.bbsrc.ac.uk
    Point your web browser at ftp://ftp.res.bbsrc.ac.uk/pub/aar/ 
    in the IACR-Rothamsted information system  http://www.res.bbsrc.ac.uk/ 

To join the asreml discussion list, send the message  
     subscribe
to  asreml-request@chiswick.anprod.CSIRO.au

The address for messages to the list is asreml@chiswick.anprod.CSIRO.au

                        <> <> <> <> <> <> <>
"Seek first the kingdom of God and His righteousness"  Jesus; Matthew 6: 33
<><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><><>