Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Felix Idelberger <felix.idelberger@neptun.uni-freiburg.de> |
To | statalist@hsphsun2.harvard.edu |
Subject | st: How Do I limit the observations used in a regression? |
Date | Sat, 14 Sep 2013 13:17:27 +0200 |
Dear Folks, It's my first time that I pose a question to this mailing list. First I would like to express my gratitude towards all that are helpful to others including me. I have about 1775 observations of which 1400 show very high values of my dependent variable. That's why I want to limit the number of observations with high values of my dependent variable which will be used in my regression. Additionally, I have 6 independent dummy-variables which should be all equally represented in the reduction of these observations to let's say 240. These 1400 observations all have a value between 90-100. Does someone know a good way to do this? I would like to tell Stata that it should only use every fifth observation that matches my criterion and that it observes an equal amount of observations that fulfill one of my dummy-variables. Thank you in advance. Yours sincerely, Felix Idelberger * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/faqs/resources/statalist-faq/ * http://www.ats.ucla.edu/stat/stata/