Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: RE: Centiles categorization of a continous variable for index creation


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: Centiles categorization of a continous variable for index creation
Date   Tue, 18 Feb 2003 20:19:38 -0000

Fernando Montenegro-Torres
>
> I am trying to create an index using based upon the score
> results from
> principal components analysis I would like to express the
> results of the
> score as an index from 0-100.
> Centiles function gives me the centiles of the score but I
> cannot find a
> way to categorize each value of the score (I only have 26
> observations)
> according to the centiles. Can someone give me a suggestion. I tried
> egen with centiles but does not work, egen does not accept centiles.

If you think a PCA is worth doing it is not clear
why you want to throw away information by mapping results
(e.g. PC1) to centiles. As any component has
arbitrary upper and lower bounds it seems that
you may just need for your purposes some MIN and
MAX in say

	PC1 - MIN
100 *	---------
	MAX - MIN

but it would seem that there is always some
arbitrariness about MAX and MIN, always assuming
that they span the data.

Alternatively if you want what are sometimes
called percentile ranks or plotting positions,
there are -egen- solutions spelled out in
detail within an FAQ at

How can I calculate percentile ranks?
How can I calculate plotting positions?
http://www.stata.com/support/faqs/stat/pcrank.html


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

*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   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   |   What's new   |   Site index