| Date: | Fri, 8 Oct 1999 11:19:18 +0200 |
|---|
| Reply-To: | Winston Groenewald <jacov@ABSA.CO.ZA> |
|---|
| Sender: | "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU> |
|---|
| From: | Winston Groenewald <jacov@ABSA.CO.ZA> |
|---|
| Organization: | Amalgamated Banks of South Africa Ltd. |
|---|
| Subject: | Re: Definition of CHAID? |
|---|
In addition, it is a technique that one uses to reduce the number of
categories of predictor variables. CHAID divides a population into two or
more distinct groups based on the best predictor of a dependent variable.
This splitting continues until there are no more statistically significant
predictors.
Winston Groenewald
Johannesburg, South Africa.
SFBAY0001 wrote in message <19991007132626.01760.00000062@ng-fv1.aol.com>...
>it stands for Chi Squared Automatic Interation Detection
>Andrew H. Karp
>Sierra Information Services, Inc.
>A SAS Institute Quality Partner in the USA
>1489 Webster Street Suite 1308 San Francisco, CA 94115 USA
>415/441-0702 (voice)
>SierraInfo@AOL.COM
>http://www.SierraInformation.com
>
| 
