Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: Interaction effects


From   Maarten buis <maartenbuis@yahoo.co.uk>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Interaction effects
Date   Mon, 14 Jun 2010 10:38:31 -0700 (PDT)

--- On Mon, 14/6/10, Lorenzo Ciari wrote:
> I am estimating a probit model and my dependent variable is
> entry (0-1) in a given market. I estimate the model using
> form level data, so for each firm I observe whether it
> enters or not a market (I have multiple markets): I want to
> test whether entry depends on a given variable (call it
> COMP) and see whether the effect of COMP on entry is
> particularly strong for firsm with certain characteristics
> (suppose the characteristics SIZE).
> 
> 1)  Question 1: if I want to test the hipothesis that
> COMP has no effect for values of SIZE lower that X, can I
> create a dummy = 1 for firms with size<X and then
> estimate a model with COMP, SIZE (continuous variable) and
> the interaction between (COMP) and the dummy? Can I
> interpret the interaction coefficient as in OLS (I wouldn't
> know how to use AI-Norton inteff command within this
> framework, as I do not interact SIZE with COMP, but the new
> created dummy....

Not really. If you want to interpret the effects in terms of
marginal effects, then you will have to find a way to make
-inteff- work for you. Alternatively, you can switch to a 
-logit- model and interpret the exponentiated coefficients,
see: <http://www.maartenbuis.nl/publications/interactions.html>

As I discuss in that paper, both are perfectly legitimate ways 
of conceptualizing interaction effects, but they measure subtly
different things.
 
> 2) Question2: can I, instead, put SIZE in my regression and
> the generate three dummies, one for low values of SIZE, one
> for medium and one for high values. Then I would interact
> COMP with the three dummies and estimate a model with three
> separate coefficients for COMP, that is COMPlow, COMPmedium
> and COMPhigh...

I don't think that this changes much, so once you have solved
your first problem, the answer to your second problem will
follow automatically. (of course you need to decide whether
you want to leave out either a reference category [COMPlow,
or COMPmedium, or COMPhigh] or leave out the main effect
[COMP], as you can't enter all four)

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/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index