Dear ASREML friends,
I am having difficulty to specify the right variance structure for a
bi-variate analysis of data where there are 10 repeated records per cow.
The Cow and anim are representing the same thing but I have included cow to
fit
premanent environmental effect. If possible at a later stage I am planning
to run
multivariate analysis of repeated observations using a random regression.
Initially I want to try a two trait repeatability model and estimate
genetic,
environmental and permanent environmental correlation between the traits.
Below is the .as and part of the .asr file. I would be very thankful if
someone could help.
Mekonnen
The as file
bivariateparity1-test-analysis
anim !I !P
cow 12218 !I
dpc
herd 395 !I
state
location 26 !A
calving
age
parity
yeart 7 !I
montht 12
dayt
testn 10
timec
lcl
tc 40 !A
fc 3
proh !*0.1
scc !^0
milk !*0.1
fat !*0.01
pr !*0.01
ped51.csv !ALPHA !REPEAT !MAKE
p51.csv !RECODE !REPORT !MAXIT 200
scc milk ~ Trait Tr.year Tr.pol(age,2) Tr.montht Tr.pol(timec,3),
!r Tr.anim Tr.ide(cow) !f Tr.herd
1 2 2 !STEP .05
12218
2 US 0.13442 0.077 9.0139
Tr.anim 2
2 0 US 0.292 0.0950 0.6109
28467 0 AINV
Tr.ide(cow) 2
2 0 US 0.1 0.02 0.4
12218 0 IDEN
The asr file
12218 identity
Fault 1 Failed to parse R structure line
scc p51.csv
Model specification: TERM LEVELS GAMMAS
Tr.pol(timec 8 0.00 Tr.montht 24 0.00 Tr.pol(age,2 6
0.00
Tr.year 14 0.00 Trait 2 0.00 Tr.anim 56934
0.10
Tr.ide(cow) 24436 0.10 Tr.herd 790 0.00
SECTIONS 12218 9 1
TYPE 0 0 0 0 0 0 0 0
STRUCT 12218 0 0 0 0 1 0 0
2 0 0 0 0 1 0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
33 factors define [max 500].
9 variance parameters [max 550]. 2 special structures
Final parameter values 0.10000 0.10000 0.0000
-.10000E-36
Last line read was: 2 US 0.13442 0.077 9.0139 0 0 0 0 0
33 1 122180244360 64000
Finished: Wed Sep 8 10:33:12 1999 Failed to parse R structure line
--
Asreml mailinglist archive: http://www.chiswick.anprod.csiro.au/lists/asreml