[R] Simple decision trees - rpart?
perbak at tdcadsl.dk
Sun Apr 30 16:59:47 CEST 2006
I'm looking for help on how to use R for making simple decision trees to
calculate EMV's, in a manner similar to the method applied in the
PrecisionTree add-in for Excel. Say, what is the optimal action, if:
- could have two out comes with probabilities of 60% and 40% and associated
values of 100 and 20
- could have three out comes with probabilities of 30%, 60% and 10% with
values of 200, 30 and -50
I have tried to read up on the 'rpart' functionality, but I think I might be
looking in the wrong direction. I would also like to work with continuous
distribution functions within the tree.
Can anyone point me in the right direction?
More information about the R-help