Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

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


From   Clare L Maxwell <[email protected]>
To   [email protected], [email protected]
Subject   Re: st: Classification analysis/risk tree (CART) in Stata - anything more?
Date   Sun, 22 Apr 2007 17:01:29 -0500

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, [email protected] 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/



© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index