I am using the rpart package and would like to know if there is a way to set
a minimum limit on how many rows there can be in a single factor to be
considered. I am getting results back on columns that have several factors,
but each factor will only have 2 or 3 rows. So it meets the minbucket and
minsplit criteria when they get grouped together. See below for an example
(leaf 6 has 12 rows, 6 factors so each one has an average of 2 actual data
points). I would like to set the cutoff above 2 to avoid this.
Thanks, Don
n=57 (200 observations deleted due to missing)
node), split, n, deviance, yval
* denotes terminal node
1) root 57 122.910600 1.5271950
2) X.1.A=A53 ,A59 ,A61 ,A63 ,A64 31 50.516670 0.9799769
4) X.2.A=A53 ,A58 ,A59 ,A60 ,A64 ,A72 16 5.221407 0.1872424 *
5) X.2.A=A56 ,A57 ,A61 ,A62 ,A63 15 24.515240 1.8255600 *
3) X.1.A=A56 ,A57 ,A58 ,A60 ,A62 ,A72 26 52.043030 2.1796470
6) X.S..A=A54 ,A57 ,A58 ,A59 ,A61 ,A62 12 9.819111 0.9266623 *
7) X.S..A=A53 ,A56 ,A60 ,A63 ,A64 14 7.236000 3.2536340 *
-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._