Resumen de Computing value correspondences for repeated games with state variables
The nature of repeated interaction has been extensively studied in the repeated garne literature. Abreu (1988), Abreu, Pearce and Stacchetti (1986, 1990), and Cronshaw and Luenberger (1994) develop a recursive approach to characterizing repeated games by focusing on the present values of subgame perfect strategies for each player, V. Judd and Conklin (1995), Cronshaw and Rutherford (1994) and Cronshaw (1996) have implemented these techniques computationally. Sorne of the most interesting examples of strategic interaction, however, arise in environments with state variables in which the recursive techniques cited aboye cannot be employed. In such environments the set of values of subgame perfect equilibrium becomes a function of the state variable-the object of interest becomes the value correspondence. This paper presents a general method for computing value correspondences under perfect monitoring and discounting.
