Essays in empirical industrial organization

In order to understand strategic interactions among firms, economists often need to structurally model and estimate games. The two essays in this dissertation develop a new method for estimating static discrete games with complete information, and present an application of such methods for a data set from the lodging industry. The first essay proposes to exploit the relationship between the set of Nash equilibria and the set of correlated equilibria to estimate such games. The second essay uses the concept of correlated equilibrium to empirically study pre-entry communication and market structure.;The estimation of static discrete games with complete information has posed fundamental challenges to economists because the possibility of multiple equilibria invalidates the use of standard approaches such as Maximum Likelihood Estimation and the General Method of Moments. The first essay proposes a two-step estimator to resolve the issue of multiple equilibria by exploiting the fact that all of the Nash equilibria are contained in the set of correlated equilibria. In the first step, I estimate the conditional choice probabilities of each possible outcome. In the second step, I obtain bounds for the parameter estimates by minimizing the average distance between the set of correlated equilibria and the probability distribution obtained in the first step. Compared to previous approaches to the estimation of games, my method has two important advantages. First, it explicitly takes into account the possible existence of mixed strategy equilibria. Second, it is computationally easy to implement: due to the inherent linearity of correlated equilibria, I can obtain the bounds estimates by solving a series of linear programming problems.;The second essay adopts a structural model to study pre-entry communication and market structure in the lodging industry using the methods I have developed in the first essay. By assuming potential entrants to play a correlated equilibrium, the model takes into account of pre-entry communication among firms, since the concept of correlated equilibrium automatically incorporates pre-play communication. I first obtain the parameter estimates of the entry game based on correlated equilibrium, and then simulate the counterfactual where each potential entrant has to make decisions simultaneously and independently, that is, by playing Nash equilibrium. The comparison across the true and counterfactual regimes suggests that pre-entry communication does not have a significant impact on the number of firms in a market. However, the communication does seem to increase within-market dispersion in capacity levels. In addition, pre-play communication increases industry profit.