Lachenbruch, Peter wrote:
>>A short question on PCA for categorical variables: wouldn't
correspondence analysis be useful here? Or is my interpretation of CA
as the categorical analog of PCA way off base?<
Yes, and no, respectively. :)
Now for something like Poisson counts, CA isn't directly the tool for the job. At this point I suspect doing something like square root transforming the counts and then using ordinary PCA would probably better but there's just no way of telling short of knowing more.
Jay
*
* 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/
