Title: Prediction Error Learning in Univariate Generalized Autoregressive Models
Abstract: The rational expectations hypothesis is supported if rational expectations are stable with respect to reasonable learning procedures. We consider within a univariate generalized autoregressive model a class of prediction error based learning procedures which contain as a special case the OLS-procedure. We derive a concept of stability, characterize stable and unstable rational expectations equilibria, determine the possible limit outcomes of the learning procedures and show that there is convergence with unit or only positive probability, depending on the respective feedback function which maps the perceived into the true law of motion of the endogenous variable. Finally, in view of the plausibility of the learning procedures, we determine the rate of convergence.
Keywords: Rational expectations, learning, model specification
JEL-Classification-Number: C22, C62, D83, D84
Creation-Date: March 1994
Unfortunately this paper is not available online. Please contact us to order a hardcopy.
SFB 303 Homepage
17.02.1998, © Webmaster