Recursive game

A stochastic game with pay-off at the end of a play (see also Dynamic game). Since a recursive game can be endless, it is essential to determine the pay-off of the players in the case of infinite plays. An analysis of any Shapley game can be reduced to an analysis of a certain recursive game, but because of the possibility of infinite plays, research on recursive games is generally more complicated than research on stochastic games. Any zero-sum two-person finite recursive game has a value and both players have stationary $\epsilon$-optimal strategies. H. Everett [1] has demonstrated a method of finding both the value of the game and of the optimal strategies.

References

[1]	H. Everett, "Recursive games" in Dresher, M. (ed.), Tucker, A. W. (ed.), Wolfe, P. (ed.), Contributions to the theory of games III, Annals of Mathematics Studies 39, Princeton Univ. Press (1957) pp. 47–87. Zbl 0078.32802 Zbl 0078.31001.

References

[a1]	S. Alpern, "Games with repeated decisions" SIAM J. Control Optim. , 26 : 2 (1988) pp. 468–477