Cournot-Nash equilibrium and optimal transport in a dynamic setting

TL;DR

This paper explores dynamic population games with evolving types, linking Nash equilibria to causal optimal transport, and introduces a numerical scheme for computing these equilibria, demonstrated through a congestion game case study.

Contribution

It establishes a novel connection between dynamic Cournot-Nash equilibria and causal optimal transport, including existence, uniqueness, and a new numerical method.

Findings

Established existence and uniqueness of equilibria in potential games.

Developed the first numerical scheme for causal optimal transport.

Demonstrated the approach with a congestion game case study.

Abstract

We consider a large population dynamic game in discrete time. The peculiarity of the game is that players are characterized by time-evolving types, and so reasonably their actions should not anticipate the future values of their types. When interactions between players are of mean-field kind, we relate Nash equilibria for such games to an asymptotic notion of dynamic Cournot-Nash equilibria. Inspired by the works of Blanchet and Carlier for the static situation, we interpret dynamic Cournot-Nash equilibria in the light of causal optimal transport theory. Further specializing to games of potential type, we establish existence, uniqueness and characterization of equilibria. Moreover we develop, for the first time, a numerical scheme for causal optimal transport, which is then leveraged in order to compute dynamic Cournot-Nash equilibria. This is illustrated in a detailed case study of a congestion game.