Singularity In Average Information Matrix Using Ainverse

From: Michelle <asremlforum_at_VSNI.CO.UK>
Date: Mon, 21 Sep 2009 22:03:21 +0100

Hi,

Thanks everyone for being so helpful on my previous questions- I just have one more that I was wondering if someone could help me out with.

I am doing a mixed model analysis and using the following R code:

MGS.ainv = asreml.Ainverse(MGS.ped, mgs=TRUE)\$ginv
attr(MGS.ainv, "rowNames")

geno\$Animal<-as.factor(geno\$Animal)
test.asr = asreml(Fert ~ SNP, random= ~ ped(Animal,var=T), ginverse=list(Animal=MGS.ainv), data=geno)

When I run the analysis I get:

asreml(): 2.00 Library: 2.00bi Run: Mon Sep 21 15:37:38 2009
Equations: 787 (2 dense)
Initial update shrinkage factor: 0.316
LogLik S2 DF

-109.5005 0.4832 571 15:37:38
-95.9616 0.4206 571 15:37:38
-81.4739 0.3275 571 15:37:38
-70.8107 0.2024 571 15:37:38
-67.0327 0.1079 571 15:37:38
-65.7958 0.0485 571 15:37:38
-65.3938 0.0168 571 15:37:38
-65.2692 0.0034 571 15:37:38 (1 component(s) constrained)
-65.2432 0.0002 571 15:37:38 (1 component(s) constrained)
-65.2416 0.0000 571 15:37:38
Warning: 1 singularities inAverage Information Matrix
Exit status: -1 - Singularity in Average Information Matrix

Finished on: Mon Sep 21 15:37:38 2009

Error: Abnormal termination

I know that others have received this error but I'm not sure I've understood the responses and have not seen anyone using ASREML-R that has gotten it. Does anyone know how to solve this problem? Any information you can provide me with would be much appreciated.

Thanks,

Michelle

-------------------- m2f --------------------

Sent using Mail2Forum (http://www.mail2forum.com).