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 on April 23, and its replacement, statalist.org is already up and running.


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

st: RE: sort of standardization


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   "Statalist" <statalist@hsphsun2.harvard.edu>
Subject   st: RE: sort of standardization
Date   Wed, 12 May 2010 20:04:39 +0100

Several answers to this all I think omitted one point about PCA. 

Standardisation in the sense of (value - mean) / sd is unnecessary before PCA because you only need ask for a PCA based on a correlation matrix. 

However, you're not guaranteed the same result with (value - min) / (max - min) OR value / (max - min). 

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

From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Ginevra Biino

I have to standardize many variables (in order to run PCA).
Besides generating the n corresponding std(varname) vars, which I have already done, I also want to generate n new  variables obtained dividing each variable by its range. Can anybody help me?

 

*
*   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