Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Christopher Baum <kit.baum@bc.edu> |
To | "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu> |
Subject | Re: st: first-difference regression |
Date | Wed, 28 Sep 2011 04:02:20 -0400 |
<> On Sep 28, 2011, at 2:33 AM, donsaane wrote: > When you run a first- difference method, do you always > put : noconstant ? For example : reg dx dy, noconstant, or > you just run it like OLS . eg : reg dx dy > Because I read an article and they put «noconstant» , so I > want to know if when we run a first difference, we always put «noconstant» If you write down a linear model in levels and take first differences, the constant disappears. If one of the regressors is a time trend, its difference is constant, and your FD regression will have a constant term. If you do not believe there should be a trend in the level equation, then you should not use a constant in the FD equation. Kit Kit Baum | Boston College Economics & DIW Berlin | http://ideas.repec.org/e/pba1.html An Introduction to Stata Programming | http://www.stata-press.com/books/isp.html An Introduction to Modern Econometrics Using Stata | http://www.stata-press.com/books/imeus.html * * 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/