Identification and Estimation of Discrete Games of Complete Information

We discuss the identification and estimation of discrete games with complete information. Following Bresnahan and Reiss, a discrete game is defined to be a generalization of a standard discrete choice model in which utility depends on the actions of other players. Using recent algorithms that compute the complete set of the Nash equilibria, we propose simulation-based estimators for static, discrete games. With appropriate exclusion restrictions about how covariates enter into payoffs and influence equilibrium selection, the model is identified with only weak parametric assumptions. Monte Carlo evidence demonstrates that the estimator can perform well in moderately-sized samples. As an illustration, we study the strategic decisions of firms in spatially-separated markets in establishing a presence on the Internet