CHAID Package -- Recursive partitioning and decision trees

[Leo Martinez]

Hello!
I am using Stata 12.1 and am a relatively new Stata user.

I have a question regarding the CHAID package. I need to produce multiple decision trees and use recursive partitioning using a cross-sectional database. I read online in one article that the CHAID package in Stata only really is allowable if the dataset is failure time data and not cross-sectional data. Since this package is fairly new however I'm wondering if this is still true. Have updates been made yet to use CHAID with cross-sectional data? If so, are there any online tutorials on it's use in Stata?


Thanks in advance!
Leo

[Nick Cox]

chaid is from SSC. Please note the request to explain where user-written programs come from, a request spelled out in the FAQ all are asked to read before posting: http://www.statalist.org/forums/help#stata

I have not used chaid but the assertion cited here (absolutely no reference for us to check) on the face of it appears completely absurd. Perhaps if context or content is added, it may appear less absurd.

The author of chaid, Joseph Luchman, is a member here. If he doesn't respond, I would advise contacting him directly.

Note: Perhaps there is some confusion with cart (SSC), a completely different program, which is restricted to failure time data.

[Leo Martinez]

Ya I apologize. I saw CART and for some reason was confused with the chaid package. I'm not sure how that happened. Forget my question.