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]

st: xtprobit / Including Interaction Effect with Constant


From   "Christian Pfarr" <christian.pfarr@uni-bayreuth.de>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: xtprobit / Including Interaction Effect with Constant
Date   Fri, 2 Mar 2012 17:06:32 +0100

Dear Stata Community,

I want to estimate a binary discrete-choice experiment with a random-effects
probit model (xtprobit) including interactions of the socio-economic
variables with the constant term. How can I generate such interaction terms?

Background: The parameter coefficients are alternative specific and
individual effects are invariant across an individual's decisions. Due to
the Random Utility Model only utility differences are relevant. Thus,
individual effects that do not vary are eliminated by the differencing.
To capture individual effects, I have to interact socio-economic
characteristics with the factor (varying across alternatives) under
consideration. This is straightforward with variables such as the price
attribute etc. However, I am not sure how to handle the constant term. Some
Studies suggest interacting socio-economic characteristics with the constant
term to capture e.g. the status quo bias. One potential solution could be to
include a dummy for each of my 56 alternatives. But this doesn't seem very
reasonable to me. Do you have any alternative suggestions?

Thank you very much for your advice.

Christian

-------
University of Bayreuth
Mail to: christian.pfarr@uni-bayreuth.de



*
*   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