This paper provides a learning justification for limited forecast equi
libria, i.e., strategy profiles such that (1) players choose their act
ions in order to maximize the discounted average payoff over their hor
izon of foresight as given by their forecasts and (2) forecasts are co
rrect on and off the equilibrium path. The limited forecast equilibria
appear to be the stochastically stable outcomes of a simple learning
process involving (vanishing) trembles. (C) 1998 Academic Press.