Solutions and properties of multi-stage stackelberg games

The inducible region concept is used to solve multistage, deterministic Stackelberg games, and to study properties of the solutions. We first delineate the inducible region by using a process similar to dynamic programming. A desired outcome is then selected from the region. To construct a Stackelberg strategy, we move forwards in time to cover various contingent situations. This approach gives a distinct picture about the nature of the problem. In particular, two aspects of the principle of the optimality are discussed. A design procedure is presented to construct a relatively robust Stackelberg strategy. The issue of credibility is also addressed.