topic Re: Silhouette coefficient in SAS Enterprise Guide

Silhouette coefficient

ryesidariza — Fri, 07 Jun 2019 19:14:26 GMT

How do I calculate the silhouette coefficient, and what's the code for that? Or an example I can follow?

Re: Silhouette coefficient

PGStats — Fri, 07 Jun 2019 20:12:41 GMT

Seems to be covered here:

https://www.sas.com/content/dam/SAS/support/en/sas-global-forum-proceedings/2019/3409-2019.pdf

Re: Silhouette coefficient

Reeza — Fri, 07 Jun 2019 20:25:44 GMT

Which tool are you using and what exact version? Some will have this and some you'll have to do your own calculations.

Re: Silhouette coefficient

ryesidariza — Fri, 07 Jun 2019 20:35:51 GMT

Thanks

Cluster k means para sas enterprise guide 7.1

Re: Silhouette coefficient

Reeza — Fri, 07 Jun 2019 20:55:58 GMT

Can you run the following and post the output from the log? SAS versioning is different than EG versioning.

proc product_status;run;

Re: Silhouette coefficient

Diego_L — Mon, 10 Jun 2019 15:01:20 GMT

proc product_status; run;

For Base Product ...
Custom version information: 9.21_M3
Image version information: 9.02.02M3P032410
For SAS/STAT ...
Custom version information: 9.22
Image version information: 9.02.02M0P033110
For SAS/GRAPH ...
Custom version information: 9.21_M2
For SAS Integration Technologies ...
Custom version information: 9.2_M2
For SAS/ACCESS Interface to PC Files ...
Custom version information: 9.21_M2

Re: Silhouette coefficient

Reeza — Mon, 10 Jun 2019 15:17:22 GMT

You're on a really old version (9.2 was released in 2008) so I doubt it has that measurement. Unfortunately that means you'll need to calculate it manually. Otherwise, try PROC SCORE or PLM for starting points.

Re: Silhouette coefficient

Diego_L — Mon, 10 Jun 2019 16:27:41 GMT

we are currently in the process of updating, from which version can I find the silhouette coefficient?
Do you have an example that I can follow in a more current version?
thank you

Re: Silhouette coefficient

RalphAbbey — Mon, 16 Mar 2020 18:02:42 GMT

To my knowledge the Silhouette Coefficient isn't calculated as part of any SAS procedure or SAS Viya action. When I wrote the SAS Global Forum paper my goal was to discuss a lot of the methods that people use to evaluate clustering, but not necessarily limit myself to methods that are supported within the software.

There are some definite down-sides to the Silhouette Coefficient in that 1) you shouldn't use it to compare two different types of clustering - it's very much related to centroid based clustering methods 2) it is a time consuming metric to calculate if you have large numbers of observations.

That being said, I think that one of the better ways to implement if you are looking to do so in SAS would be by using SAS IML. Because the Silhouette Coefficient looks across all pairs of observations the matrix setup of IML would be an easier programming approach for this type of problem (as compared to Data Step - which is still doable).

Re: Silhouette coefficient

RalphAbbey — Wed, 17 May 2023 15:11:00 GMT

Rick Wicklin has recently written a wonderful 2-part blog post about this.

https://blogs.sas.com/content/iml/2023/05/15/silhouette-statistic-cluster.html

https://blogs.sas.com/content/iml/2023/05/17/compute-silhouette-sas.html

Hopefully this can help.