Re: st: Classification analysis/risk tree (CART) in Stata - anything more?

[Clare L Maxwell <maxwellcl1@earthlink.net>]

Hello, Nicola.

Thank you, but no, actually, I do not have failure time data, but 
cross-sectional data, subjects whom I would like to categorize into 
risk groups to target for potential intervention.  The methodology 
sought comes from "the monograph by Breiman, Friedman, Olshen and 
Stone" mentioned by Nick below.

At 6:52 PM +0200 4/22/07, nicola.baldini2@unibo.it wrote:

> I can't understand from your post if you are aware of -cart- (module 
> to perform Classification And Regression Tree analysis for failure 
> time data. It uses the martingale residuals of a Cox model to 
> calculate approximate  chisquare values for all possible cutpoints 
> on all the CART covariates) from ssc and are searching for something 
> better, or you simply don't know about it.
>
> Nicola
>
> At 02.33 22/04/2007 -0400, Clare L Maxwell wrote:
>
> > Has Stata thought about this any more?  Especially, might CART be a
> > possibility in Stata within the next year?
> >
> > Retrieved from www.stata.com/statalist/archive/2003-04/msg00018.html
> >
> > Re: st: RE: Classification analysis / risk tree
> >
> > Subject Re: st: RE: Classification analysis / risk tree
> > Date Tue, 01 Apr 2003 06:14:06 -0600
> >
> > Paul Seed said:
> >
> > Does anyone know of a CART-type package for Stata.
> >
> >
> > Nick Cox replied:
> >
> > CART -- usually I think regarded as an abbreviation of
> > Classication and Regression Trees, after the monograph
> > by Breiman, Friedman, Olshen and Stone -- has been requested
> > several times as an addition to official Stata, including
> > at the NASUG meeting in Boston on March 17. The Stata
> > Corp response is in essence "We'll think about it".
> >
> > Fred Wolf added:
> >
> > Some listers might be interested in the commercial package called
> > "CART" and sold by Salford Systems at a reasonable cost. It has been
> > the major CART package for almost 2 decades. Recently, CART has been
> > seen as a "data mining" tool by Salford systems. This has led to
> > increases in prices to levels in excess of thousands of dollars for
> > new purchases as well as to timed licenses. The next most convenient
> > approach to CART is through S-plus.
> >
> >
> > I add:
> >
> > In her review with example in"Classification and Regression Tree
> > Analysis in Public Health:  Methodological Review and Comparison with
> > Logistic Regression," Stephanie Lemon, et al, uses "The C&RT analysis
> > was conducted using Answer Tree 2.0, a product of SPSS."
> >
> > But I'd like to use Stata.
> >
> > Thank you.
> >
> >                                 Clare Maxwell
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/