Statalist The Stata Listserver


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

Re: Re: st: "testing" a cluster analysis


From   khigbee@stata.com
To   statalist@hsphsun2.harvard.edu
Subject   Re: Re: st: "testing" a cluster analysis
Date   Wed, 07 Feb 2007 08:40:10 -0600

Ronan Conroy <rconroy@rcsi.ie> replying to Adam Seth Litwin gives
good warning about F tests after a cluster analysis

> Careful. An analysis of variance is a hypothesis test. The model is  
> specified in advance and the anova calculates the values of the model  
> parameters.
> 
> In your case, the model was generated from the data. The usual  
> interpretation of the F ratio does not apply.

You may also want to look at the -cluster stop- command with the
-calinski- option.  It provides the Calinski/Harabasz Pseudo-F
index.  The word "Pseudo" is used to emphasize the point that
Ronan makes above.

The usual use of -cluster stop- is to compare the four group
versus five group versus six group cluster solutions (as an
example) for the clustering of a dataset so that you can decide
the most natural number of groups.

Ken Higbee

*
*   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 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index