Matlab Implementation of the AIM Algorithm: A Beginner's Guide

The Anderson-Moore algorithm provides a well-established solution method for forward-looking linear rational expectations models. It is widely used at the Federal Reserve Board for a variety of purposes, ranging from simulations of macroeconometric models to computations based on models of monetary policy. The aim of this paper is to support a wider use of the Anderson-Moore method by discussing the practical sides of its application. I describe the features of one of its Matlab implementations that is freely downloadable from the web. Experience shows that one is usually required to spend quite some time in order to fully understand how the available Matlab functions work. The úemphasis is on the structures that should be modified to tailor the programs to one's needs. I also present the application of the algorithm to Coenen and Wieland (2000)'s macromodel of the Euro area.