Adaptive Learning in Practice

42 Pages Posted: 18 Jul 2006

See all articles by Eva Carceles-Poveda

Eva Carceles-Poveda

SUNY at Stony Brook University - College of Arts and Science - Department of Economics

Chryssi Giannitsarou

University of Cambridge - Faculty of Economics; Centre for Economic Policy Research (CEPR)

Date Written: April 2006

Abstract

We analyse some practical aspects of implementing adaptive learning in the context of forward-looking linear models. In particular, we focus on how to set initial conditions for three popular algorithms, namely recursive least squares, stochastic gradient and constant gain learning. We propose three ways of initializing, one that uses randomly generated data, a second that is ad-hoc and a third that uses an appropriate distribution. We illustrate, via standard examples, that the behaviour and evolution of macroeconomic variables not only depend on the learning algorithm, but on the initial conditions as well. Furthermore, we provide a computing toolbox for analysing the quantitative properties of dynamic stochastic macroeconomic models under adaptive learning.

Keywords: Adaptive learning, least square estimations, computational methods, short-run dynamics

JEL Classification: C63, D83, E10

Suggested Citation

Carceles-Poveda, Eva and Giannitsarou, Chryssi, Adaptive Learning in Practice (April 2006). CEPR Discussion Paper No. 5627. Available at SSRN: https://ssrn.com/abstract=916569

Eva Carceles-Poveda

SUNY at Stony Brook University - College of Arts and Science - Department of Economics ( email )

Stony Brook, NY 11794
United States

Chryssi Giannitsarou (Contact Author)

University of Cambridge - Faculty of Economics ( email )

Austin Robinson Building
Sidgwick Avenue
Cambridge, CB3 9DD
United Kingdom

Centre for Economic Policy Research (CEPR)

London
United Kingdom

Register to save articles to
your library

Register

Paper statistics

Downloads
19
Abstract Views
650
PlumX Metrics