Empirical Games: The Oligopoly Case

In this paper, I examine methods used to construct and estimate game-theoretic models of oligopolistic interactions, and I survey some of the empirical findings. Both static and dynamic games are covered. Static models provide useful summary statistics concerning the outcomes of oligopolistic interaction, whereas dynamic models attempt to capture more complex strategic behavior. In the exposition of static games, I consider two empirical approaches: the first-order approach and the dual approach. When we turn to dynamics, I limit my remarks to state-space games - games in which actions taken in one period shift payoffs in subsequent periods. With state-space games, all history that is pay-off relevant is summarized by a small set of variables - the state - and players' strategies depend only on the state and time.