As the subject says, I am unhappy with R's aov performance. I have a
data
set containing 25000 cases. This causes thrashing even with very
moderate
formulae, because the model matrix has quite a lot of lines.
The study has 9x3x2x2x2 (or so) design factors. Is there a recommended
method for pre-condensing the data before inputting them into aov in R?
I want to be able to preserve the person factor (42) for further
analysis.
I could try to aggregate the response variables for all and use a count
for weighting them.
AFAIS, in principle the cases could be submitted seqentially to the
fitting
process avoiding the explicit building of the model matrix. Is anybody
working
on this in R context?
--
Dipl.-Math. Wilhelm Bernhard Kloke
Institut fuer Arbeitsphysiologie an der Universitaet Dortmund
Ardeystrasse 67, D-44139 Dortmund, Tel. 0231-1084-257
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !) To: r-help-request at
stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._
