In this paper we present a general single-agent stochastic dynamic opt
imization model that is capable of dealing with learning and with rati
onal expectations. We obtain the solution by solving at every time ste
p the rational expectations model and then applying existing optimizat
ion schemes to derive the optimal solution.