Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | Ridhima Gupta <g.ridhima@gmail.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | Re: st: marginal effect of the interaction term in the probit model using margins command |
Date | Thu, 30 Jun 2011 12:00:45 +0100 |
Thanks a lot Maarten. Your comments helped to put things in perspective. On 6/30/11, Maarten Buis <maartenlbuis@gmail.com> wrote: > On Thu, Jun 30, 2011 at 12:08 PM, Ridhima Gupta <g.ridhima@gmail.com> wrote: >> But Maarten doesn't this criticism apply to marginal effects in >> general and not just the marginal effect on the intearction term? > > Marginal effects of main effects are often close enough to constant > and they hardly ever switch signs. This is not true for marginal > effects of interaction terms. I think of marginal effects as a linear > model fitted on top of our non-linear model. Models are wrong, but as > long as they are close enough, they can be useful. Average marginal > effects for regular variables are often close enough so they can be > useful. For interaction effects that linear model of the non-linear > model just breaks down. > > -- 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/ > * * 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/