R help - Feb 2002 - question regarding to The tree Package for R

I have a problem with running the tree package (dec.8, 2001) for R.  The
problem is,
it will only give me 5/6 terminal node and then stop, while using Splus's
tree on the
same data with the same specification give me hundreds of nodes.

Here's a little more background info:

    R-1.4.1
    Solaris 5.7
    rpart  (most recent version)
    tree   (..)

    Splus 6.0
    Solaris 5.7
    tree

    dataset:
        100,000 records
        about 3 variables
        regression tree

    data read in from flat file,
    summary() on R and Splus showed identical distributions

    tree were run with all the deault setting (no mincut, minsize set etc)

    with added options: minsize=2, mindev=-1,  get the same small tree.

Did I miss anything?

thanks

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

Hello,

Recently it was posted in this Help for a similar question that the package
to
use is "rpart" in any case, it seems that in your case the
configuration
parameters
for the tree are different in the R version you have. You can set up several
different
parameters in "tree" to define which are the minimum number of
observations
to
partition. If I do not remember wrongly, in S-Plus this number is 5.

Carlos Ortega.


-----Mensaje original-----
De: owner-r-help at stat.math.ethz.ch
[mailto:owner-r-help at stat.math.ethz.ch]En nombre de Huayin Wang
Enviado el: jueves, 21 de febrero de 2002 20:19
Para: r-help at stat.math.ethz.ch
Asunto: [R] question regarding to The tree Package for R


I have a problem with running the tree package (dec.8, 2001) for R.  The
problem is,
it will only give me 5/6 terminal node and then stop, while using Splus's
tree on the
same data with the same specification give me hundreds of nodes.

Here's a little more background info:

    R-1.4.1
    Solaris 5.7
    rpart  (most recent version)
    tree   (..)

    Splus 6.0
    Solaris 5.7
    tree

    dataset:
        100,000 records
        about 3 variables
        regression tree

    data read in from flat file,
    summary() on R and Splus showed identical distributions

    tree were run with all the deault setting (no mincut, minsize set etc)

    with added options: minsize=2, mindev=-1,  get the same small tree.

Did I miss anything?

thanks

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.
-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._.
_._


_____
The information in this email is confidential and it may not be
disclosed or used by anyone other than the addressee. If you are not the
intended recipient, any disclosure, copying, distribution or any action taken or
omitted is prohibited and may be unlawful.
Minorplanet cannot accept responsibility for the accuracy or completeness of
this email as it has been transmitted over a public network. If you suspect
that the email may have been amended, please call the sender.


-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._

I was having the same problem.  Try adding "mindev = 0" in your tree()
command.  That fixed it for me.

HTH

--------------------------------------------------------------------------
Stephen W. Juranich                             sjuranic at ee.washington.edu
Electrical Engineering             http://students.washington.edu/sjuranic
University of Washington                http://ssli.ee.washington.edu/ssli


 /"\                          || 
 \ /  ASCII RIBBON CAMPAIGN   ||  NO ATTACHMENTS
  X   AGAINST HTML MAIL       ||  NO STATIONERY
 / \  AND POSTINGS            ||  NO GRAPHICS



-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
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
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._