Re: [R] interperting a regression tree

From: <Bill.Venables_at_csiro.au>
Date: Fri, 21 Mar 2008 15:40:22 +1000

If you have fitted a tree model, the diagram (which did not survive the mailing list, of course) would have been a decision tree, not a dendrogram.

In a decision tree, the internal nodes refer to decisions to be made as to whether to proceed to the left or to the right at that stage. This is based on the value of a selected predictor variable. If tm is the tree model, you would normally display it using

plot(tm)
text(tm)

Each internal node is then labelled with a message line "age < 25.5", which shows that at that point the decision is made on the variable, 'age'. If the age of the case is less than 25.5 choose the left branch, otherwise the right branch.

When you arrive at a terminal node, bingo! You get a prize.

Bill Venables

-----Original Message-----
From: r-help-bounces_at_r-project.org [mailto:r-help-bounces_at_r-project.org] On Behalf Of Sharma, Manju
Sent: Thursday, 20 March 2008 8:54 PM
To: r-help_at_r-project.org
Subject: [R] interperting a regression tree

hello
i am trying to decipher a dendrogram i have from performing a 'tree'. (attached file)
my response variable is factored - low, medium and high threat and the 8 explanatory variables are numeric values.  

i could do with some help to understand what the values <0.25, <0.5 and <0.75 are on the tree branches? how does one interpret this???  

cheers,
mann      

Manju Sharma
Ashoka Trust for Research in Ecology and the Environment, Banaglore, India   



R-help_at_r-project.org 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. Received on Fri 21 Mar 2008 - 05:47:14 GMT

Archive maintained by Robert King, hosted by the discipline of statistics at the University of Newcastle, Australia.
Archive generated by hypermail 2.2.0, at Fri 21 Mar 2008 - 08:30:23 GMT.

Mailing list information is available at https://stat.ethz.ch/mailman/listinfo/r-help. Please read the posting guide before posting to the list.

list of date sections of archive