Stochastic policy design in a learning environment with rational expectations

In this paper, we present a method for using rational expectations in a sto chastic linear-quadratic optimization framework in which the unknown parame ters are updated through a learning scheme. We use the QZ decomposition as suggested by Sims (Ref. 1) to solve the rational expectations part of the m odel. The parameter updating is done with the Kalman filter and the optimal control is calculated using the covariance matrix of the uncertain paramet er.