Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Franco Solleza <solleza.f@gmail.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | st: Fixed Effects with ID specific Time Trends - Computation Time |
Date | Wed, 6 Mar 2013 10:33:17 -0500 |
Hi, I am currently running a regression on panel data where the right hand side includes fixed effects on an id variable, and a time trend for each id variable. As you might imagine this leads to a large number of parameters (twice the number of id variables minus 1 -- for the dummy). Ignoring for a moment issues that might arise concerning degrees of freedom, the issue I am running into is when there are many id variables. Sometimes, the regressions have ~28k parameters. The command I am using to run this model is: xi: areg y x id*_time_trend [aw=weightvariable], absorb(id) The issue I run into is that these models take more than 24 hours to run! I've been trying to figure out a way to speed up computation time but have not been able to do so. If anyone has any suggestions on how to speed things up, that would be great! Thanks! Best, -- fps solleza.f@gmail.com * * 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/