Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | "Martin Weiss" <martin.weiss1@gmx.de> |
To | <statalist@hsphsun2.harvard.edu> |
Subject | st: AW: RE: Filtering data |
Date | Wed, 30 Jun 2010 12:40:04 +0200 |
<> " I note that discarding 20% of your data is a pretty drastic solution to whatever it is that you see as the problem." I remember there being criticism at last year`s UK UGM <http://www.stata.com/meeting/uk09/abstracts.html> of a presenter who omitted certain extreme cases of his dependent variable from his estimation. The details of the incident have become somewhat hazy for me, though... HTH Martin -----Ursprüngliche Nachricht----- Von: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] Im Auftrag von Nick Cox Gesendet: Mittwoch, 30. Juni 2010 11:39 An: statalist@hsphsun2.harvard.edu Betreff: st: RE: Filtering data Martin Weiss has already answered. On a different level, I note that discarding 20% of your data is a pretty drastic solution to whatever it is that you see as the problem. Transformations? Non-identity link functions? Nick n.j.cox@durham.ac.uk Maximiliano Manuel Silva Correa Im currently analizing travel times for serveral urban bus trips in the city of Santiago, Chile. I'm using lots of data coming from GPS sources. Pretty newbie to Stata, Im stuck trying to filter data by bus trip route codes; Id like to drop all observations wich have travel times under percentile 10 and over percentile 90 for all observations sharing a certain route code... this filter for all route codes. * * 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/ * * 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/