Agglomeration Table for Cluster Analysis

[Adam Guerrero]

Hi All,

It is possible to produce agglomeration schedules after a hierarchical agglomerative cluster analysis in Stata?

A PowerPoint presentation shared at the 4th German Stata Users Group meeting alludes to commands that perform this task, but I am having a hard time locating the installations. A link to the presentation can be found on Stata's website:

http://www.stata.com/meeting/4german/abstracts.html

A link to the presentation is below the abstract for "New tools for evaluating the results of cluster analyses," by Hildegard Schaeper.

The command syntax used to generate the agglomeration schedule is

cluster stop [clname], rule(schedule) [laststeps(#)]

Does anyone know where I can download the package that includes this command?

Thanks so much in advance,

Adam

[Brendan Halpin]

As far as I can tell the agglomeration schedule is in the _clus_N_hgt variable.

Something like this should work:

gsort -_clus_1_hgt
gen increase = _clus_1_hgt - _clus_1_hgt[_n+1]
list _clus* increase in 1/15,clean

If you want the Adjusted Rand Index, it's in my SADI package: ssc install sadi
See ari, and also permtab

[Adam Guerrero]

Thank you so much for your reply, Brendan. I will definitely experiment w/ those commands. Thank you so much for sharing.

Best,
Adam