In Game Theory, we generally refer to "normal form" and "extensive form" as representations. And, we generally describe "Nash Equilibrium," "strictly dominated strategies," "maxmin strategies," "correlated equilibrium," "correlated rationalizability" and "self-conforming equilibrium" as solution concepts. Yet, some scholars refer to "Bayesian Games" or "Repeated Games" as representations while others refer to them as game types. What then are the analytical definitions for a "solution concept," "representation," and "game type?" And, are "Bayesian Games" and "Repeated Games" better defined as representations, game types, or something else altogether?

  • $\begingroup$ From an analytical perspective, it seems that one could make the following distinctions in game theory: 1) Basic Elements (e.g., Players; Outcomes); 2) Mathematical Representations (e.g., Normal Form); 3) Solution Concepts (e.g., Correlated Equilibrium); 4) Game Types (e.g., Repeated Games; Bayesian Games; Conditional Games); 5) Standard Cases (e.g., Prisoner's Dilemma). Thoughts? $\endgroup$ – EVolk Dec 16 '15 at 6:21
  • $\begingroup$ I'm not sure there are standard analytical definitions for most of those terms. Osborne and Rubinstein are usually precise on a variety of terms, such as equilibrium and game form. $\endgroup$ – Trurl Dec 24 '15 at 16:01

Thanks to Harsanyi, Bayesian games can be represented as an extensive form game with Nature (or chance) moving at the initial node. So I'd say they are more of a game type: games with incomplete information.

Repeated games are a special case of extensive form games. So they are also a game type, I'd say.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.