Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

RE: st: How do I test that two subsample have different coefficient of variation?


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   RE: st: How do I test that two subsample have different coefficient of variation?
Date   Fri, 11 Jul 2008 17:18:59 +0100

I don't doubt you can often get away with the cv when the implicit
assumptions are not too unrealistic on the whole. The same applies to
almost all statistical methods. 

Yes, the Kelvin scale plays some part in climatology, and could be used
more, but it's not conventional for the contexts I was quoting. 

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

Austin Nichols

Nick--
Suppose the original poster wants to compare income distributions--the
sd of log income is a common measure applied to such distributions,
but income can be zero or negative, though mean income in some
population is unlikely to be zero. The CV is a natural comparison
measure, as is squared CV or any other inequality measure discussed in
Stephen Jenkins' talk
(http://www.stata-journal.com/article.html?article=st0095) or SSC
programs (findit jenkins).

ps. why not degrees Kelvin?

On Fri, Jul 11, 2008 at 5:57 AM, Nick Cox <n.j.cox@durham.ac.uk> wrote:
> There are some references in
>
> Sokal, R.R. and Rohlf, F.J. 1995. Biometry. New York: W.H. Freeman.
>
> The rough argument for thinking logarithmically goes like this. It
makes
> sense to work with the coefficient of variation whenever standard
> deviation is proportional to mean. That implies that variability is
> multiplicative, not additive, which in turn implies working on a
> logarithmic scale.

*
*   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