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

st: hadimvo: multivariate outlier detection


From   "G. ter Riet" <G.terRiet@amc.uva.nl>
To   statalist@hsphsun2.harvard.edu
Subject   st: hadimvo: multivariate outlier detection
Date   Mon, 11 Aug 2003 13:57:22 +0200

Dear List members,

Could anyone tell me what the interpretation of the hadi-distance is? 
The Stata manual (see hadimvo) states that it is "the distance (not the 
distances squared) from the basic subset", but I do not understand 
this. Perhaps because I expected that I could interpret these distances 
as z-values, but playing around a bit with some data created using the 
commands:

-corr2data sbp age sex com1, n(1000) m(120 50 .4 .2) sd(12 
7 .048989 .04)-
AND
-hadimvo sbp age sex com1, gen(odd distance)-
AND
-su  distance-

showed distances varying between 0.227 and 4.429, and not values 
roughly between -3 and 3.
Any help will be greatly appreciated.

Gerben ter Riet, MD PhD, Academic Medical Center
Dept. General Practice, Room J3-354, 1105 AZ Amsterdam

*
*   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