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

st: RE: Chernoff's faces


From   "Nick Cox" <n.j.cox@durham.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: Chernoff's faces
Date   Wed, 30 Oct 2002 12:12:34 -0000

Matthias  Greiner
> 
> I wonder whether someone has written a Stata programme for graphing
> Chernoff's faces. grstar is nice but faces are somewhat 
> easier to interpret
> (or not?).

-findit- finds nothing. I have 
not heard of anything in Stataland. 
Also, I think this would be quite 
difficult to program in Stata. 

My impression is that these caught a lot 
of attention for a few years as a very 
original idea, but have not really been adopted
anywhere as a standard method. That's partly chicken 
and egg: only if Chernoff faces were implemented,
would people have a chance to play, and I don't 
think they have been widely implemented. 

A common remark is that they worked 
quite well with some non-statistical 
audiences, perhaps for poor reasons 
(novelty, cutiness, ...), but statistically minded 
people tend to be critical of 
arbitrariness of coding and difficulties
of interpretation, and to suspect that 
e.g. cluster structure, outliers, 
etc. are usually best detected in 
other ways, say on some 2-D projection(s) 
or a tree structure. 

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