Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Mauro Mastrogiacomo <M.Mastrogiacomo@cpb.nl> |
To | "'statalist@hsphsun2.harvard.edu'" <statalist@hsphsun2.harvard.edu> |
Subject | RE: st: Year Fixed Effect Interpretation |
Date | Thu, 8 Sep 2011 10:22:13 +0200 |
One other alternative is the method of Angus Deaton & Christina Paxson, 1998. "Saving and growth: another look at the cohort evidence," Working Papers 225, Princeton University. They use a linear transformation of the time dummies (dropping 2), which somehow goes back to Maarten's suggestion. In this way they manage to include at the same time, age time and cohort effects, which are otherwise linearly related. Mauro -----Original Message----- From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Maarten Buis Sent: donderdag 8 september 2011 10:10 To: statalist@hsphsun2.harvard.edu Subject: Re: st: Year Fixed Effect Interpretation On Wed, Sep 7, 2011 at 10:18 PM, Venkiteshwaran, Vinod wrote: > I am running a pooled regression with dummies(t-1 dummies) as controls for year effects. I have chosen to drop the first year so that I can interpret the coefficients on the remaining year dummies in terms of the first year. At this stage I have to include an additional independent variable that is collinear with these year dummies so Stata automatically drops another dummy , for the last year, before running the regression. My focus is on whether or not the coefficients on the dummy variables systematically increase or decrease over time. However, since the dropped dummies are at the beginning and end of the sample the trend in the dummy coefficient is no longer present. I have tried to drop two year dummies at the beginning of the sample but still the time trend seems to have vanished! Sorry, what you want to do is impossible. The fact that your year dummy is perfectly collinear with your covariate means your data does not contain the necessary information to get separate estimates of both variables. So part of the effect of your covariate should be included in your time trend, but there is no information in your data to determine which part or how large that part would be. One way to solve this is to add time is linear curve or maybe a spline rather than as a set of dummies. Hope this helps, Maarten -------------------------- Maarten L. Buis Institut fuer Soziologie Universitaet Tuebingen Wilhelmstrasse 36 72074 Tuebingen Germany http://www.maartenbuis.nl -------------------------- * * 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/ -- ================================================================================ Dit bericht kan informatie bevatten die niet voor u is bestemd. Indien u niet de geadresseerde bent of dit bericht abusievelijk aan u is toegezonden, wordt u verzocht dat aan de afzender te melden en het bericht te verwijderen. De Staat aanvaardt geen aansprakelijkheid voor schade, van welke aard ook, die verband houdt met risico's verbonden aan het elektronisch verzenden van berichten. This message may contain information that is not intended for you. If you are not the addressee or if this message was sent to you by mistake, you are requested to inform the sender and delete the message. The State accepts no liability for damage of any kind resulting from the risks inherent in the electronic transmission of messages. ================================================================================ * * 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/