|
Learning is a new way of thinking about macroeconomic dynamics. It was used initially to assess the plausibility of rational expectations equilibria (REE) and to select among equilibria when there are multiple REE. Lately, more work focuses on empirical performance and policy implications. The modern approach to adaptive learning differs from the adaptive expectations approach of the 1950s by using rational expectations equilibria as a reference point and by emphasizing the implications of small deviations from full forecast rationality. The macroeconomic predictions induced by adaptive learning dynamics recently allowed interpreting and replicating a variety of empirical phenomena that would otherwise appear puzzling from the viewpoint of RE models.
This course reviews the implications of relaxing the rational expectations assumption in dynamic models. We focus on applications to macroeconomics and asset pricing. We study the implications of replacing the rational expectations hypothesis by the view that agents are learning and constantly trying to improve their forecasts. The course reviews the basic theoretical results of the learning literature, applications of learning models to explain empirical phenomena and to inform the design of fiscal and monetary policy.
|
Home
