This award funds research in the theory of games with incomplete information. This branch of noncooperative game theory has a wide range of applications in Economics and other social sciences. Examples include the analysis of optimal taxation, auctions and procurement design, signaling models of the labor market, adverse selection models of trade and insurance markets, reputation and commitment in long-run relationships, credibility of monetary policies, and strategic trading in financial markets.
Previous research has established that the predictions made by game theory about behavior in these situations are quite sensitive to the assumptions made about the players' infinite hierarchies of beliefs. The PIs develop new methods to study this robustness problem. They evaluate the severity of the robustness problem by characterizing necessary and sufficient conditions for the strategic behavior of a given Harsanyi type to be approximated by the strategic behavior of (a sequence of) perturbed types. This amounts to providing a belief-based characterization of the strategic topology over types. The investigators also apply their characterization to a variety of questions concerning robustness to perturbations of higher-order beliefs, including genericity of common priors and of critical types, and the connections between robustness of strategic behavior and the notion of common p-belief.
The project will have broader impact because this branch of game theory is used to study applied problems in multiple domains, including Computer Science, and Biology. The PIs will be incorporating their research into graduate courses and will train graduate students in these new methods.
