gameL8 - Extensive Games with Imperfect Information In...

Info iconThis preview shows pages 1–6. Sign up to view the full content.

View Full Document Right Arrow Icon
Extensive Games with Imperfect Information In strategic games, players must form beliefs about the other players’ strategies, based on the presumed equilib- rium being played. In Bayesian games, players must form beliefs about the other players’ strategies and their types, based on the probability distribution over types and the presumed equi- librium being played. In extensive games with perfect information, there is the possibility that a player will face a situation that is in- consistent with the presumed equilibrium being played. However, subgame perfection takes care of this issue by requiring a form of sequential rationality, even o f the equilibrium path.
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
In extensive games with imperfect information, when a player faces a situation that is inconsistent with the pre- sumed equilibrium being played, she may also be forced to form beliefs about the other players’ past behavior. These beliefs are often crucial in evaluating whether the ensuing play is rational. Put another way, games with imperfect information may have few subgames, possibly only one. How then are we to eliminate Nash equilibria that involve threats that are not credible?
Background image of page 2
De f nition 200.1: An extensive game consists of the following components: 1. The set of players (assumed to be a f nite set), N . 2. A set, H , of sequences (histories of actions) satisfying the following properties: (2a) The empty sequence, ,isane lemento f H . (2b) If ( a k ) k =1 ,...,K H ,(whe re K may be in f nity) and L<K ,then ( a k ) k =1 ,...,L H . (2c) If an in f nite sequence, ( a k ) k =1 ,..., satis f es ( a k ) k =1 ,...,L H for every positive integer L ,t h e n ( a k ) k =1 ,..., H . 3. A function, P , that assigns to each non-terminal history a member of N { c } .( P is the player func- tion, with P ( h ) being the player who takes an action after the history, h .I f P ( h )= c holds, then chance or nature takes the action after the history, h .A h i s t o r y , ( a k ) k =1 ,...,K ,is terminal if it is in f nite or if there is no a K +1 such that ( a k ) k =1 ,...,K +1 H .)
Background image of page 3

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
4. A function, f c , that associates with each h H for which P ( h )= c , a probability measure on A ( h ) , denoted by f c ( ·| h ) . Eachsuchmeasureisassumedto be independent of every other such measure. 5. For each player, i N , a partition = i of { h H : P ( h i } with the property that A ( h A ( h 0 ) when- ever h and h 0 are in the same element of the partition. For I i = i ,wedeno teby A ( I i ) the set A ( h ) and by P ( I i ) the player P ( h ) for any h I i . 6. For each player, i N , a preference relation º i on lotteries over the set of terminal histories, Z .
Background image of page 4
I i = i is called an information set of player i .Ap l a y e r cannot distinguish two histories (nodes) in one of her information sets. Notice the requirement that the set of available actions must be the same for any two histories in an information set, or else the player would be able to distinguish between the histories.
Background image of page 5

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
Image of page 6
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 07/17/2008 for the course ECON 817 taught by Professor Peck during the Fall '07 term at Ohio State.

Page1 / 21

gameL8 - Extensive Games with Imperfect Information In...

This preview shows document pages 1 - 6. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online