Re: [R] Regression tree: labels in the terminal nodes

[Achim Zeileis]

On Thu, 16 Aug 2007, Juergen Kuehn wrote:


Dear everybody,


I'm a new user of R 2.4.1 and I'm searching for information on improving
the output of regression tree graphs.

In the terminal nodes I am up to now able to indicate the number of
values (n) and the mean of all values in this terminal node by the command

> text(tree, use.n=T, xpd=T)


Note that this is specific to the "rpart" implementation. There are also 
other recursive partitioning algorithms available, see e.g., the 
"MachineLearning" task view at

  http://CRAN.R-project.org/src/contrib/Views/MachineLearning.html

The ctree() algorithm in package "party" provides more flexibility 
concerning visualization. The plotting methods are all extensible although 
this is not extensively documented.


Maybe you also want to look at Simon Urbanek's interactive KLIMT software 
for tree visualization.


hth,
Z


Yet I would like to indicate automatically in the output graph of the
tree some quality measure, e.g. impurity (or standard deviation .)
and a character to identify which cases are in one terminal node, e.g.
given a ID number or name.

Until now I did not discover in my R help scripts or in the R programm
help how to do it. So I calculate impurity by hand, and I'm identifying
the cases in each node "by hand", and I am adding these values with a
graphical software to the graph (as e.g. given "*.jpg" file). Therefore
anybody can see that I added these values by hand. I'm quite sure that
there is a more easy and faster way which looks (and is!) more professional.

Could anybody help me? That would be great!


Thank you very much for your support!


Dr. Jürgen Kühn
Leibniz-Centre for Agricultural Landscape Research (ZALF)
Institute of Soil Landscape Research_
_
Eberswalder Str. 84
D-15374 Müncheberg
Tel.: ++49/(0)33432/82-123
Fax: ++49/(0)33432/82-280
E-mail: [EMAIL PROTECTED] 
Internet: http://www.zalf.de/home_zalf/institute/blf/blf/

__
R-help@stat.math.ethz.ch 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.

__
R-help@stat.math.ethz.ch 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.

[R] Regression tree: labels in the terminal nodes

[Juergen Kuehn]

Dear everybody,


I'm a new user of R 2.4.1 and I'm searching for information on improving 
the output of regression tree graphs.

In the terminal nodes I am up to now able to indicate the number of 
values (n) and the mean of all values in this terminal node by the command

 > text(tree, use.n=T, xpd=T)

Yet I would like to indicate automatically in the output graph of the 
tree some quality measure, e.g. impurity (or standard deviation .) 
and a character to identify which cases are in one terminal node, e.g. 
given a ID number or name.

Until now I did not discover in my R help scripts or in the R programm 
help how to do it. So I calculate impurity by hand, and I'm identifying 
the cases in each node "by hand", and I am adding these values with a 
graphical software to the graph (as e.g. given "*.jpg" file). Therefore 
anybody can see that I added these values by hand. I'm quite sure that 
there is a more easy and faster way which looks (and is!) more professional.

Could anybody help me? That would be great!


Thank you very much for your support!


Dr. Jürgen Kühn
Leibniz-Centre for Agricultural Landscape Research (ZALF)
Institute of Soil Landscape Research_ 
_
Eberswalder Str. 84
D-15374 Müncheberg
Tel.: ++49/(0)33432/82-123
Fax: ++49/(0)33432/82-280
E-mail: [EMAIL PROTECTED] 
Internet: http://www.zalf.de/home_zalf/institute/blf/blf/

__
R-help@stat.math.ethz.ch 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.