Learnability of E-Stable Equilibria

Posted: 22 Dec 2013

See all articles by Atanas Christev

Atanas Christev

Heriot-Watt University; IZA Institute of Labor Economics

Sergey Slobodyan

Charles University in Prague - CERGE-EI, a joint workplace of Charles University and the Economics Institute of the Czech Academy of Sciences; HSE Moscow

Date Written: November 26, 2012

Abstract

If private sector agents update their beliefs with a learning algorithm other than recursive least squares, expectational stability or learnability of rational expectations equilibria (REE) is not guaranteed. Monetary policy under commitment, with a determinate and E-stable REE, may not imply robust learning stability of such equilibria if the RLS speed of convergence is slow. In this paper, we propose a refinement of E-stability conditions that allows us to select equilibria more robust to specification of the learning algorithm within the RLS/SG/GSG class. E-stable equilibria characterized by faster speed of convergence under RLS learning are learnable with SG or generalized SG algorithms as well.

Keywords: Adaptive Learning, Expectational Stability, Stochastic Gradient, Speed of Convergence

Suggested Citation

Christev, Atanas and Slobodyan, Sergey and Slobodyan, Sergey, Learnability of E-Stable Equilibria (November 26, 2012). Available at SSRN: https://ssrn.com/abstract=2370242

Atanas Christev (Contact Author)

Heriot-Watt University ( email )

Riccarton
Edinburgh EH14 4AS, Scotland EH14 1AS
United Kingdom

IZA Institute of Labor Economics

P.O. Box 7240
Bonn, D-53072
Germany

Sergey Slobodyan

Charles University in Prague - CERGE-EI, a joint workplace of Charles University and the Economics Institute of the Czech Academy of Sciences ( email )

Politickych veznu 7
Prague, 111 21
Czech Republic

HOME PAGE: http://www.cerge-ei.cz

HSE Moscow ( email )

26 Shabolovka
Moscow
Russia

Do you have negative results from your research you’d like to share?

Paper statistics

Abstract Views
552
PlumX Metrics