Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: Clustering variables instead of cases


From   Alexey Shpenev <shpenevstatalist@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: Clustering variables instead of cases
Date   Wed, 3 Oct 2012 22:26:38 +0400

I am performing some analyses on biomedical data. I have a set of patients with a lot of components in their blood measured and quantified. I am not interested in classifying those components.
I know there's a user-written command clv for such purposes, but I am not sure if anything better exists in stata or in some outsource packages. 
What I need is not only the results of clustering, but also some criteria on how to choose the best number of clusters and how to test those clusters for validity. I would also like to know if there are some measures that allow to test how likely each single variable is correctly classified
I have seldom used cluster or factor analysis before, so please forgive if I ask for something obvious.
Maybe you can just suggest some stata commands and I will read the help files.

Thanks a lot,
Alexey


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


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