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]

Re: st: PCA with unbalanced data


From   Stas Kolenikov <skolenik@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: PCA with unbalanced data
Date   Tue, 5 Apr 2011 21:01:00 -0500

On Tue, Apr 5, 2011 at 3:05 PM, PINAR ERDEM <fpinar_erdem@yahoo.com> wrote:
> I want to use PCA (principal componets analysis) with a dataset of 49 variables. However my data is unbalanced (unequal number of observations). My questions are if it is possible to run PCA with unbalanced data and how to get longest possible components/factors? Any suggestions would be very much appreciated.

I think this is a perfect case for -mi impute mvn-. PCA kind of
assumes normal data, so -mi- will come up with an appropriate
estimator of the covariance matrix of the data; with some luck, you
might even be able to pull it out of the guts of -mi- and analyze as
is. If not, you can impute a few dozen times... and then get stuck, as
-mi estimate- does not support -pca-.


-- 
Stas Kolenikov, also found at http://stas.kolenikov.name
Small print: I use this email account for mailing lists only.

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


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