|
Activity Number:
|
278
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Tuesday, August 4, 2009 : 8:30 AM to 10:20 AM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
| Abstract - #304227 |
|
Title:
|
Bayesian Inference for Semiparametric Dynamic Choice Quantal Response Equilibrium Models
|
|
Author(s):
|
Anton Westveld*+ and Kevin Quinn
|
|
Companies:
|
University of Nevada Las Vegas and Harvard University
|
|
Address:
|
Dept of Mathematical Sciences, Las Vegas, NV, 89154-4020,
|
|
Keywords:
|
Bayesian semiparametrics ; multi-agent choice ; quantal response equilibrium ; social interaction data
|
|
Abstract:
|
Building on work of McKelvey and Palfrey (1995, 1996, 1998) and Signorino (1999) on quantal response equilibrium models and Newton et al. (1996); Ishwaran and Zarepour (2000); Ishwaran and James (2001) and Ishwaran and Zarepour (2002), on semiparametric Bayesian methods, we develop Bayesian inference procedures for a particular class of semiparametric strategic choice models. We model the stochastic portion of an actor's random utility function through an approximation to the centrally standardized Dirichlet process prior of (Newton, Czado & Chappell 1996). Model ?tting is accomplished via Markov chain Monte Carlo. We present results from Monte Carlo experiments and a simulated data example. Our Monte Carlo results show that incorrectly specifying the distribution of the actors' utility disturbances within a parametric model can dramatrically bias quantities of interest.
|
