Deterministic Equilibrium Selection Under Payoff-Perturbed Dynamics

39 Pages Posted: 6 Jul 2014

See all articles by Ryota Iijima

Ryota Iijima

Yale University, Law School, Students

Date Written: May 16, 2014

Abstract

This paper studies models of population game dynamics where players make stochastic choices because of payoff perturbations. The goal is to obtain deterministic equilibrium selection, where the action distribution in the population globally converges to Nash equilibria with probability 1 in a finite time, under a moderate level of payoff perturbations. Exploiting the tractable nature of bounded perturbations, we characterize almost-sure global stability in a finite population setting by generalizing risk-dominance in many action games. We also characterize conditions for (approximate) global asymptotic stability under a continuum population setting allowing unbounded perturbations; this leads to a stronger conclusion that waiting times are uniformly bounded in the population size.

Suggested Citation

Iijima, Ryota, Deterministic Equilibrium Selection Under Payoff-Perturbed Dynamics (May 16, 2014). Available at SSRN: https://ssrn.com/abstract=2462656 or http://dx.doi.org/10.2139/ssrn.2462656

Ryota Iijima (Contact Author)

Yale University, Law School, Students

127 Wall Street
New Haven, CT 06511
United States

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

Paper statistics

Downloads
141
Abstract Views
1,088
Rank
370,717
PlumX Metrics