Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


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

st: RE: PCA with Missing Values (or other factor analysis)


From   Nick Cox <n.j.cox@durham.ac.uk>
To   "'statalist@hsphsun2.harvard.edu'" <statalist@hsphsun2.harvard.edu>
Subject   st: RE: PCA with Missing Values (or other factor analysis)
Date   Tue, 5 Jul 2011 19:48:27 +0100

Essentially any statistical command can only use observations that are all non-missing. 

Either you live with that or you need to tell us why and how it is reasonable to impute votes not made by people who were not in office. I doubt there's a third choice. 

The difference between PCA and factor analysis is immaterial here. By the way, I don't know that many recent texts regard PCA as a kind of factor analysis, although this point of views was common (e.g.) in the 1960s. 

Nick 
n.j.cox@durham.ac.uk 

Nick Eubank

I'm trying to do a PCA on legislator votes (1 or 0), where different
legislators have overlapping but distinct terms. In other words,
legislator A is in office periods 1 and 2, legislator B is in office 2
and 3, so A has no votes in period 3 and B has no votes in period A. I
want to map their "ideal points" (i.e. pca predicted value from votes)
into the same space. But PCA obviously dislikes missing values.

Suggestion? I'm really open to any factor analysis algorithm.


*
*   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 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index