Statalist


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

st: RE: Cluster analyis on hand made distance matrix


From   "Verkuilen, Jay" <JVerkuilen@gc.cuny.edu>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: Cluster analyis on hand made distance matrix
Date   Mon, 10 Mar 2008 12:11:27 -0400

Ulrich Koehler wrote:

>However, if I use wards-linkage I get different results for the two distance matrizes: [snip]<

You may have run into a local optima problem. Clustering methods and other multivariate data reduction procedures are plagued by this. The clustering program should give you the option of starting from different configurations, possibly many random starts or from a given starting value.  I don't know enough about Ward's linkage to know if is is more susceptible to this problem than the others.   

Jay

<<winmail.dat>>




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