Learning and equilibrium in misspecified models

18 Pages Posted: 30 Oct 2019 Last revised: 30 Jun 2020

See all articles by Filippo Massari

Filippo Massari

University of Bologna, Department of Economics

Jonathan Newton

Kyoto University - Institute of Economic Research

Date Written: June 30, 2020

Abstract

We consider learning in games that are misspecified in that players are unable to learn the true probability distribution over outcomes. Under misspecification, Bayes rule might not converge to the model that leads to actions with the highest objective payoff among the models subjectively admitted by the player. From an evolutionary perspective, this renders a population of Bayesians vulnerable to invasion. Drawing on the machine learning literature, we show that learning rules that outperform Bayes’ rule suggest a new solution concept for misspecified games: misspecified Nash equilibrium.

Keywords: misspecified learning, evolutionary models

JEL Classification: D8, C7, C4

Suggested Citation

Massari, Filippo and Newton, Jonathan, Learning and equilibrium in misspecified models (June 30, 2020). Available at SSRN: https://ssrn.com/abstract=3473630 or http://dx.doi.org/10.2139/ssrn.3473630

Filippo Massari (Contact Author)

University of Bologna, Department of Economics ( email )

Bologna
Italy

Jonathan Newton

Kyoto University - Institute of Economic Research ( email )

Yoshida-Honmachi
Sakyo-ku
Kyoto 606-8501
JAPAN

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

Paper statistics

Downloads
182
Abstract Views
956
Rank
301,572
PlumX Metrics