There is indeed a version for R somewhere. I think it is discussed
briefly in Venables and Ripley's book. Without having used
either I guess that the commercial CART has more bells
and whistles, but otherwise the R version is distinctly
(even infinitely) cheaper.
I guess a general Stata version is unlikely in finite time. CART
is manifestly not trivial, while at the same time I sense that -- for
slightly different reasons -- that both Stata developers
and user-programmers have little interest in devoting the time
to writing an implementation, especially given the R implementation.
Naturally, one counter-example would smash this guess.
Nick
[email protected]
Fred Wolfe
> In the 1980s and early 1990s the CART program (Salford Systems) was
> relatively inexpensive. Recently (when I last looked several
> years ago) its
> price was in the thousands of dollars and the license was not
> perpetual.
> It's marketed as "data mining" to corporate users.
>
> S-Plus has a Cart program and a presume that the free R also has it.
> >I am trying to do a CART-type analysis, but the only
> >relevant package I have found, van Putten's -cart-
> >package, is of no use for me.
> >Are there any other cart packages available for stata?
> >If not, would the people who has done this type of
> >analysis please point in the direction of the
> >statistical software they used?
*
* 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/
