Statalist


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

Re: st: How to drop some Observation in Stata, if they are out of range?


From   Maarten buis <maartenbuis@yahoo.co.uk>
To   stata list <statalist@hsphsun2.harvard.edu>
Subject   Re: st: How to drop some Observation in Stata, if they are out of range?
Date   Tue, 23 Sep 2008 10:09:47 +0100 (BST)

--- bumbuminc wrote to me privately:
> Thanks for your answer, but the problem is i don't now the min and
> max values. I want to drop the observations if they are under 2.5%
> and over 97.5% of the maximun earning.

Please don't respond off list, the reason for that is discussed in the
statalist FAQ point 5 (a link to the statalist FAQ can be found at the
bottom of this and every other message on the statalist).

I think such procedures are a bad idea. If you are worried about
outliers look at the appropriate diagnostic graphs (see: -help regress
postestimation-) and identify individual influential observations and
try to ascertain why they are outliers, then use that information to
decide what to do with that observation. 

-- Maarten

-----------------------------------------
Maarten L. Buis
Department of Social Research Methodology
Vrije Universiteit Amsterdam
Boelelaan 1081
1081 HV Amsterdam
The Netherlands

visiting address:
Buitenveldertselaan 3 (Metropolitan), room N515

+31 20 5986715

http://home.fsw.vu.nl/m.buis/
-----------------------------------------


      
*
*   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   |   What's new   |   Site index