Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Flayols Alexandra <flayols@univ-tln.fr> |
To | Statalist <statalist@hsphsun2.harvard.edu> |
Subject | st: Subsamples with fairlie method |
Date | Fri, 21 Jun 2013 12:10:34 +0200 |
Dear statalisters,I use the fairlie command available for STATA users to decompose the probability of being employed with a probit model for which I use pooled coefficient. The sample size of my two groups are differents. To address this problem I would like applied the methodology proposed by Fairlie about subsamples (2006, "An extension of the Blinder-Oaxaca decompositition to logit and probit models"). That is to say draw a random subsample of my larger group equal in size to the full other group sample. If I'm not mistaken the fairlie command proposed by Ben Jann doesn't allow to draw a large number of random subsample as he proposed in its article. Please could you indicate me if it's possible to create random subsample using this command or if I need to change the ado file ?
Thank you for your help. -- Flayols Alexandra Laboratoire d'économie appliquée au développement (LEAD), http://lead.univ-tln.fr Université du Sud Toulon -Var Faculté de Sciences Économiques et de Gestion * * 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/