Least mean squares learning in self-referential linear stochastic models

Abstract We analyze Self-Referential Linear Stochastic models under bounded rationality assuming that agents update their beliefs by means of the Least Mean Squares algorithm. This learning mechanism is less complex than Recursive Ordinary Least Squares learning and appears to be more plausible as a learning device for economic agents. We prove convergence of the learning mechanism, the convergence conditions are different from those required by Recursive Ordinary Least Squares learning.