Coordination in Games

Event details

When

Jun 18, 2013
from 05:30 PM to 06:30 PM

Where

Ken Edwards Building, Lecture Theatre 1

Contact Name

Contact Phone

0116 252 2320

Add event to calendar

Professor Chris Wallace

Department of Economics

Lecture summary


Game theory is the (mathematical) study of strategic interaction. Economists and game theorists have, by and large, reached a consensus on what kind of behaviour ought to be predicted in simple one-shot interactions (so-called "strategic games"): the actions of the agents should form a Nash equilibrium. However, there is no guarantee that any given game has a single Nash equilibrium —quite the reverse: most games of interest to social scientists typically will have many many  equilibria. In such instances the question is not whether the agents will play a Nash equilibrium in the first place, but rather on which one of the many available they will coordinate.

Traditional game-theoretic approaches (for instance, the vast "refinements" literature of the 1970s and 80s) have not been able to offer a satisfactory solution to the problem of coordination in games.
Rather, it was inspiration from the evolutionary games literature in theoretical biology that provided a potential answer to this question. Stochastic evolutionary dynamics, developed by game theorists in economics in the late 1980s and early 1990s allow the analyst to "select" a single equilibrium in some of the most interesting coordination games. Such selection admits an easy interpretation in terms of the risks facing the agents making their choices in the game.

However, it soon became clear that the arguably ad hoc modelling assumptions of these early attempts to choose between equilibria were critical to the selection criterion. A more convincing approach needed a careful micro-foundation for the specific stochastic evolutionary dynamic employed. One such approach—introducing the requisite randomness through the agents' own preferences—not only (largely) reinforces the central message of the earlier work in stochastic evolutionary games, but also allows the analysis of larger and more interesting games. Several of these games lie at the very core of the discipline: the study of the production and provision of public goods is something with which every undergraduate economics student should be familiar. Stochastic evolutionary approaches offer new and complementary insights into the reasons for the success or failure of general collective actions of this type.

Share this page:

Filed under:
Time and place

Please check the start time and location of any lecture you want to attend as these vary.

Campus maps

Contact

If you have any queries about any of these events, please contact inaugural@le.ac.uk