A nonlinear incentive strategy for multi-stage Stackelberg games with partial information

In this paper we apply a kind of non-linear incentive strategy to solve the multi-stage Stackelberg games with partial information. This strategy depends on a parameter p, as p=??, it becomes the affine strategy. In addition, by using this strategy, we can calculate the quantity of threat and we can show that the duration of punishment is only on those stages where the follower plays nonoptimally. We give illustrative examples and compare the quantity of threat in the case of partial information with that in the perfect information case.