X-No-archive: yes

Dear all,
is it a bad idea to use the algorithm implemented in the SPSS two-step
cluster analysis for the approximation of an appropriate cluster
number prior to use this cluster number in k-means? K-means perfoms
better with my data than the CF tree in the two-step analysis.
Sincerely, Carl
.
.
=================================================================
Instructions for joining and leaving this list, remarks about the
problem of INAPPROPRIATE MESSAGES, and archives are available at:
.                  http://jse.stat.ncsu.edu/                    .
=================================================================

Reply via email to