st: Package -chaid- now available on SSC

[Joseph Luchman <jluchman@gmail.com>]

With many thanks to Kit Baum, package -chaid- has been uploaded to the
SSC archive.

The -chaid- command implements Kass' (1980) CHi-square Automated
Interaction Detection (i.e., CHAID) algorithm.   CHAID is a decision
tree or machine learning algorithm for exploratory data analysis (or
data mining) that recursively searches through possible splits in the
data to uncover an optimal "decision tree" that explains the dependent
variable.  CHAID is also used for clustering or segmentation purposes
as it divides the estimation sample into discrete groups based on the
decision tree results and is a "part prediction - part clustering"
algorithm.

The Stata implementation of CHAID has been checked against the results
from the R package CHAID and produces similar, but not always
identical, results as it is implemented somewhat differently both in
terms of deriving the chi-square values as well as in terms of
stopping rules for tree branching "growth."  The Stata version of
CHAID is comparable to the R version in terms of run time (although is
somewhat slower).

To install, write -ssc install chaid-

-chaid- requires Stata 12.1

- Joseph

Reference

Kass, G. V. (1980). An exploratory technique for investigating large
quantities of categorical data. Applied statistics, 119-127.

Joseph Nicholas Luchman, M.A.
----
Behavioral Statistics Lead | Fors Marsh Group
Email: jluchman@forsmarshgroup.com
forsmarshgroup.com
----
Doctoral Candidate
Industrial Organizational Psychology
George Mason University
https://www.researchgate.net/profile/Joseph_Luchman/
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/