Massimo Castro wrote:
>>As promptly suggested (thanks Carlo!), I better add some more
information regarding the Quantal Response Equilibrium application I am
dealing with.<<
I haven't read any of this stuff in a while but seem to recall that
McKelvey's QRE was a multinomial logit model of some sort. It probably
has some funny stuff due to the dependent nature of game theory models
and thus will require a mixed specification. If so, you will have to use
GLLAMM or else switch to a multivariate normal model and use multinomial
probit, assuming the subjects have a finite choice set. (Alternatively,
look at proc mdc in SAS.) I toyed around with McKelvey's Gambit software
to solve games by QRE a few years ago but this wasn't a parameter
estimation method.
I'd check with people doing behavioral game theory to see what they're
using.
Jay (not really a behavioral decision making person but can, and, in
fact did once, play one on TV...)
--
J. Verkuilen
Assistant Professor of Educational Psychology
City University of New York-Graduate Center
365 Fifth Ave.
New York, NY 10016
Email: [email protected]
Office: (212) 817-8286
FAX: (212) 817-1516
Cell: (217) 390-4609
*
* 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/
