Mostafa Beshkar wrote:
>>I think I should explain my question in more detail, since I think
there has been some misunderstanding.
Indeed.
>>I want to estimate the following probability model (this comes from my
game-theoretic model):
Pr(s=1|X)=Pr(p>BX)
Pr(s=0|X)=1-Pr(p>BX)
where s is a binary variable, X is the vector of observable variables, B
is the vector of parameters to be estimated, and p is an unobservable
random varibale that is distributed according to F on the interval
[0,1].<<
Unless I'm mistaken, this is just an ordinary binary regression. You
actually observe S = 0 or 1, you have a vector of predictors for
characteristics of each choice. The usual random utility formulation
sets BX on the real line and uses a link function generated by making
assumptions about the distribution of the disturbance in a random
utility model. Kenneth Train's most excellent book on discrete choice
(see http://elsa.berkeley.edu/books/choice2.html) explains things quite
well.
Depending on your design, you will have dependency among observations
because you have observed choices for two players in the same game, you
are in a more complex situation requiring simultaneous equations with a
non-recursive model. There is a literature on econometrics in the
context of game theoretic models; I am aware of it but don't know much
about what's going on currently. I'm guessing that biprobit in Stata
(generalizes probit to two simultaneous equations) would be of help.
Googling and a trip to Jstor gives: Estimation of Econometric Models of
Some Discrete Games, Peter Kooreman, Journal of Applied Econometrics,
Vol. 9, No. 3. (Jul. - Sep., 1994), pp. 255-268.
I must confess this is getting out of my area.... I is just a poor, dumb
psychometrician. :)
Jay
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/