[R-sig-ME] Mixed model with a blocking factor and a nested structure inside
Aitor Larrañaga
aitor.larranagaa at ehu.es
Mon Mar 7 15:12:33 CET 2011
Dear list members,
I am trying to model in R an experiment with the BACI structure, that is, consisting in a Before-After factor and a Control-Impacted one, both fixed. I had the same sampling structure in four locations, what makes location a random blocking factor. That would be easy for me if I had not also some samplings Before (2 samplings) and After (3 samplings) the affection we generated in our systems.
In short, I have a random blocking factor, two fixed factors (BA and CI) and a random sampling factor nested within BA (BA/sampling). The number of samples within each cell is ten and I have not empty cells. The sources of variation I am interested in the model are:
-BA
-CI
-BA*CI
-BA/sampling
-CI*(BA/sampling)
I have tried many different types of syntaxes and none gives me a valid result. The packages I have used are nlme and lme4, with some preference to the first as p values are obtained more easily. One of the syntaxes I have used:
model<-lme(Response~BA*CI+BA/Sampling+CI*BA/Sampling,random=~1|Block),
gives the following error:
Error in MEEM(object, conLin, control$niterEM) :
Singularity in backsolve at level 0, block 1
Below I send an example of a dataset similar to what I am using (with just two cases per cell and three blocks for brevity).
Block BA CI sampling ResponseVariable
a B C 1 3.4
a B C 1 2.5
a B C 2 7.8
a B C 2 6.5
a A C 3 3.4
a A C 3 1.2
a A C 4 0.9
a A C 4 2.3
a A C 5 2.2
a A C 5 5.6
a B I 1 8.9
a B I 1 9.0
a B I 2 1.4
a B I 2 1.4
a A I 3 2.7
a A I 3 4.4
a A I 4 4.0
a A I 4 3.7
a A I 5 2.2
a A I 5 1.1
b B C 1 7.0
b B C 1 6.9
b B C 2 5.6
b B C 2 3.4
b A C 3 2.3
b A C 3 1.1
b A C 4 1.8
b A C 4 9.6
b A C 5 5.4
b A C 5 6.6
b B I 1 1.0
b B I 1 7.7
b B I 2 5.9
b B I 2 3.3
b A I 3 3.2
b A I 3 2.1
b A I 4 0.9
b A I 4 4.3
b A I 5 7.7
b A I 5 8.7
c B C 1 6.5
c B C 1 3.8
c B C 2 3.4
c B C 2 0.4
c A C 3 6.8
c A C 3 4.4
c A C 4 9.6
c A C 4 6.5
c A C 5 6.1
c A C 5 7.7
c B I 1 9.9
c B I 1 7.6
c B I 2 5.6
c B I 2 3.4
c A I 3 2.6
c A I 3 1.7
c A I 4 2.8
c A I 4 6.7
c A I 5 5.6
c A I 5 7.8
I would be eternally grateful to the list if someone can lead me to the answer of the matter.
All the best,
Aitor Larrañaga
More information about the R-sig-mixed-models
mailing list