This year's course will be conducted by Professor Peter Howitt of Brown University. The course will introduce students to the literature on adaptation and learning in macroeconomic systems. The central question of this literature is how an economic system as a whole behaves when the individuals in the system are not endowed with enough information to undertake a coordinated set of actions of the sort that could reasonably be modeled as a rational-expectations equilibrium. In concrete terms, this encompasses such issues as the stability of full employment equilibrium, the effects of price-flexibility on macro-stability, the role of monetary and fiscal policy in allowing the economy to converge to an equilibrium, the stability of "sunspot" and "animal-spirits" equilibria, the dynamic selection of equilibrium in cases of multiple equilibria arising from coordination problems or other sources, and the evolution of money and other central institutions of a market economy.

One objective of the course is to examine in detail several macroeconomic issues on which adaptive methods shed quite a different light than does a rational-expectations-equilibrium approach. One example is the efficacy of interest-rate targeting as a technique of monetary control. The rational expectations equilibria that exist under this kind of policy regime are in many ways superior to those that would exist under a regime of monetary targeting, or nominal-income targeting. Yet from an adaptive point of view the interest-rate-targeting regime raises a host of stability issues that do not exist under these other regimes.

Another objective is to study different techniques for modeling adaptation and learning. These include Bayesian learning, inductive learning, fictitious play, genetic algorithm and genetic programming, neural networks, stochastic approximation, routine-based learning, population dynamics, social learning, and non-parametric econometrics.

The course will consist of 3 hours of formal lectures per day for two weeks, supplemented by guest lectures, tutorials and laboratory instructions. Exercises will be assigned to give the student hands-on experience using adaptive techniques and addressing substantive macroeconomic issues. Tutorials will be conducted in the computer lab, where students will learn to implement the techniques using the numerical software package Matlab. Although experience with Matlab is not necessary, some familiarity with programming will be an advantage.

Guest lecturers will include Professor Masanao Aoki of the University of California, Los Angeles, Professor Jasmina Arifovic of Simon Fraser University, Vancouver Canada, Dr. Robert Axtell of the Brookings Institution, Washington DC, Professor Richard Day of the University of Southern California, Los Angeles, and Professor Seppo Honkapohja of the University of Helsinki, Finland.

A reader will contain various articles to be studied. Secondary texts include Bounded Rationality in Macroeconomic, by Thomas Sargent (Oxford, 1993), Complex Economic Dynamics, Vol II: An Introduction to Macroeconomic Dynamics, by Richard Day (MIT Press, 1999) and Learning and Expectations in Macroeconomics, by Evans and Honkapohja (Princeton University Press, 2001) A useful website is http://www.econ.iastate.edu/tesfatsi/ace.htm