> > From: Vincenzo Matassa <s185152@student.uq.edu.au>
> To: Arthur Gilmour <gilmoua@ornsun.agric.nsw.gov.au>
> Subject: ASREML Augmented Designs problem
>
> Dear Asremlers,
> I am baffled in another problem of mine. Pretend we have an
> Augmented randomised complete block experiment. That is a randomised
> complete block design with unreplicated genotypes augmenetd in the
> design. For instance:
>
> Rep1 Rep2
>
> check 5
> 1 check
> 2 6
> 3 7
> 4 check
> check 8
>
> Now if we think of the linear model;
>
> Yield= grand_mean + Block effects + Genotype +Check + Error (1)
>
> were all main effects are considered fixed.
>
> The Fitted value for say Genotype (1) should be;
>
> Yield(1)=rep1 effect + raw genotype(1) yield value (2)
>
> If I fit
>
> Yield ~ mu rep check genotype (3)
>
> in ASREML.The fitted value for Genotypoe (1) say is just the raw yield
> value. I believe it should be the form given in eq.(2).
>
QUESTION 1: how is 'genotype' coded in the CHECK plots?
If we assume genotype is 0 for check plots
and CHECK is coded a particular value for all TEST plots.
Without constraints or random effects
rep_1 effect is zero;
check_1 effect is zero and
genotype_1 effect is zero
However; if CHECK is coded 0 for all TEST plots,
then NO singularity occures in the GENOTYPE factor
so Genotype_1 will be the deviation from the mean.
X matrix becomes
mu r1 r2 C g1 g2 g3 g4 g5 g6 g7 g8
1 1 0 1 0 0 0 0 0 0 0 0
1 1 0 0 1 0 0 0 0 0 0 0
1 1 0 0 0 1 0 0 0 0 0 0
1 1 0 0 0 0 1 0 0 0 0 0
1 1 0 0 0 0 0 1 0 0 0 0
1 1 0 1 0 0 0 0 0 0 0 0
1 0 1 0 0 0 0 0 1 0 0 0
1 0 1 1 0 0 0 0 0 0 0 0
1 0 1 0 0 0 0 0 0 1 0 0
1 0 1 0 0 0 0 0 0 0 1 0
1 0 1 1 0 0 0 0 0 0 0 0
1 0 1 0 0 0 0 0 0 0 0 1
S S
Here, the first REP effect is singular and the first genotype will
be singular becasue the Check and Genotype combine to make MU
If terms were fitted in a different order in ASREML; so that MU and
REP_1 were singular, then you would get the effect you describe.
CHeck the .sln file to see what is singular.
Arthur
> I am most grateful for this discussion group. I thank you, once again for
> your advice.
>
> Regards Vince
>
>