Hi all
I'm using a regression tree program, Guide
http://www.stat.wisc.edu/~loh/guide.html
I had a couple of questions, which I hope are okay to ask on this list.
1. When writing code for the data in Guide, you have to identify whether
variables are these:
b Categorical variable that is used both for splitting and for node
modeling. It is transformed to 0-1 dummy variables for node modeling.
c Categorical variable. It is used for splitting only, i.e., no 0-1 dummy
variable transformation is performed.
f Numerical variable used only for fitting the linear models in the nodes
of the tree. It is not used for splitting the nodes.
n Numerical variable used both for splitting the nodes and for fitting the
node models.
s Numerical-valued variable only used for splitting the nodes. It is not
used as
a regressor in the linear models. This role is suitable for ordinal
categorical variables if they are given numerical values that reflect the
orderings.
I've been reading the manuals from the webpage, and I'm having some
difficulty locating the documentation about this. Anyone have any clear
idea of what these variable types are? Anywhere else I can look?
2. Part of the Guide output is in LaTeX or Lex. I guess this is a graphical
representation of the tree output. Is there an easy way to read this? I
don't use LaTeX. I just use the basic MS office stuff, on Windows XP. If I
install LaTeX, is it easy to read stuff? I also found these other programs
which are supposed to help, and am looking through them now.
Latexeditor at latexeditor.org
and
pdflatex (for changing the output to pdf)
I'll post a summary of responses (although it may take me a couple of
days).
Thanks
Gene
Gene Shackman, Ph.D.
Evaluation and Analysis Unit
NYS DOH - Division of Nutrition
150 Broadway, FL6 West
Albany NY 12204
518 402 7304 gxs03 at health.state.ny.us
---------------------------------------------------
Applied Sociologist - Finding Solutions For Society
---------------------------------------------------
|