EC742/792

Readings for linear rational expectations models lectures

There is an immense literature on solving linear rational expectations models. In this pair of lectures, I will give an overview of the theory and the computational approaches in the literature.  Although I am not a computational expert, I have worked in this area and will use some of my own papers as these allow for consistency of style etc.

There is also a large variety of computer programs that are available for this purpose. Every author cited below has made computational programs available, as have many others. The best general purpose website is DYNARE (see link below), but I will provide an introduction to programs that I have written with Mark Watson.  Part of the reason for this is that these programs have proved to be reliable in extremely large systems, while others have not always proved so. And I understand the computational issues involved in the KWRE (King-Watson Rational Expectations) approach.

On Tuesday, I will discuss the theory of LRE using the * references from the Theory section below and the following slides PPT

On Thursday, I will discuss solving some RE models using the KWRE package based on the algorithm described in KW2002 (under System reduction algorithms below) and, tangentially, some other approaches. I will post some presentation materials later, that will parallel the solution code ZIP and part of a user's manual PDF.

Since this class involves international finance/open economy macro students, there are two observations. First, the development of singular model methodology was stimulated in part by difficulties in applying straight forward reduction methods from early RBC theory (see the discussion in KW2002). Second, international models will be one application that we will consider, using some documents and programs written by Anthony Landry for past generations of EC792 students.  An overview of the two models that Anthony constructs are is contained in "Solving international real business cycle models" PDF.  His programs for the two models ZIP worked when I ran them: I did have to edit the "setpath.m" file to reflect the location of my programs, but it was smooth sailing after that. 

 

Background

Any discussion of RE models has to start with two articles. If you have not read them, you should do so.

John F. Muth, “Rational Expectations and the Theory of Price Movements,” Econometrica, vol. 29, no. 3 ( July1961), pages 313-335. PDF

Robert E. Lucas, Jr., “Econometric Policy Evaluation: A Critique,” in ThePhillips Curve and Labor MarketsPDF

I usually start by discussing these papers first. Accordingly, the first 20 slides of this PDF presentation are good background to both theory and computational parts of the lectures, but will not be explicitly covered.

Theory

* Blanchard, O.J. and C.M. Kahn, "The Solution of Linear Difference Models Under Rational Expectations," Econometrica 1980.  PDF

Watson, M.W., "Recursive solution methods for dynamic linear rational expectations models", Journal of EconometricsVolume 41, Issue 1May 1989, Pages 65-89 PDF

* King, R. G. and M.W. Watson, "The Solution of Singular Linear Difference Systems Under Rational Expectations," International Economic Review, 39 (4), November 1998. PDF

Binder, Michael & Pesaran, M Hashem, 1997. "Multivariate Linear Rational Expectations Models: Characterization of the Nature of the Solutions and Their Fully Recursive Computation," Econometric Theory, Cambridge University Press, vol. 13(6), pages 877-88, December.

System reduction algorithms

* King, R.G. and M.W. Watson, System Reduction and Model Solution Algorithms for Singular Linear Rational Expectations Models, in Computational Economics, 20 (1-2), October 2002, 57-86. PDF

Algorithms based on the QZ decomposition

* Klein, P., "Using the generalized Schur form to solve a multivariate linear rational expectations model," Journal of Economic Dynamics and Control, Elsevier, vol. 24(10), September 2002, 1405-1423. [programs available in GAUSS and MATLAB, via the internet] PDF

Sims, C.A., “Solving Linear Rational Expectations Models", Journal of Computational Economics,20(1-2), 2001, p.1-20. [programs on his Princeton website].

 

An alternative algebraic approach

 

*Gary Anderson and George Moore. "A Linear Algebraic Procedure for Solving Linear Perfect Foresight Models." Economics Letters, 17, 1985. PDF

 

Websites:

 

AIM (the Anderson-Moore approach developed at the FRB)

 

http://www.federalreserve.gov/Pubs/oss/oss4/aimindex.html

 

DYNARE (more than just linear models, also perturbation methods and estimation algorithms)

 

http://www.cepremap.cnrs.fr/dynare/

 

UHLIG TOOLKIT

Harald Uhlig has produced a " "A Toolkit for Analyzing Nonlinear Dynamic Stochastic Models Easily"

http://www2.wiwi.hu-berlin.de/institute/wpol/html/toolkit.htm