I am running version 2.15.2 64 bit version on 64 bit Windows 7. I have a
data set with the following structure:
Fixed Effect: locationFact
Random Effects: datefact, timefact nested in datefact, interactions of
datefact and timefact with locationFact
I fit the model with the latest version of lme4.
The formula is: Thick2 ~ locationFact + (1 | datefact) + (1 |
datefact/timefact) + (1 | locationFact:datefact) + (1 |
datefact/locationFact:timefact)
Other elements of output object are:
Linear mixed model fit by REML
Random effects:
Groups Name Variance Std.Dev.
locationFact:timefact:datefact (Intercept) 34.614 5.8833
timefact:datefact (Intercept) 96.795 9.8385
locationFact:datefact (Intercept) 56.375 7.5083
datefact (Intercept) 20.341
4.5101
datefact (Intercept) 20.338
4.5098
datefact (Intercept) 20.340
4.5100
Residual 447.252 21.1483
I tested this model using rmel with another software package and found
that the datefact variance is the sum of the 3 datefact variance
estimates above. Is there a way to specify the model so I do not get the
3 datefact estimates? The same applies to the datefact BLUPs. I have to
add them to get the actual BLUP.
Thank you,
Roy Robertson
______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
