06.MixedStrategy

# 06.MixedStrategy - MIXED STRATEGIES A PURE STRATEGY is...

This preview shows pages 1–4. Sign up to view the full content.

MIXED STRATEGIES A PURE STRATEGY is a (sure) complete, contingent plan of action (which is to be followed with probability one). A MIXED STRATEGY is a probability distribution over the set of pure strategies (assigns a probability to each pure strategy). Example : if the set of pure strategies is { s 1 , s 2 , s 3 , s 4 }, a mixed strategy could be { (0.25)s 1 , (0.25)s 2 , (0.5)s 3 , (0)s 4 }. This player plans to play s 1 with probability 0.25, s 2 with probability 0.25 and s 3 with probability 0.5. He or she, however, will eventually be choosing s 1 , or s 2 or s 3 . HOW TO FIND NASH EQUILIBRIA IN MIXED STRATEGIES? Example 1: The inspector and the thief The inspector and the thief play a simultaneous-move game where the inspector chooses between “ inspect ” and “ don’t inspect ,” and the thief chooses between “ rob ” and “ don’t rob .” Thief Inspector Rob don’t inspect 10 , 0 0 , 5 don’t 0 , 20 5 , 5 1

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

View Full Document
This game has no Nash equilibrium in Pure Strategies. A mixed strategy for the Inspector assigns a probability distribution over his set of pure strategies , {inspect, don’t inspect} (0.25) Inspect Mixed strategy (0.75) Don’t inspect Play “ inspect ” with 25%, “ don’t inspect ” with 75% probability. CALCULATION OF EXPECTED PAYOFFS: Suppose there is no “outside ” uncertainty (no “moves” by Nature) 1. Given a PURE strategy profile {s 1 , s 2 }, the outcome is completely determined, hence also the payoffs: Player 1 gets U 1 (s 1 , s 2 ), and Player 2 gets U 2 (s 1 , s 2 ). Example: U Inspector (Inspect , Rob) = 10, U Thief (Inspect , Rob) = 0 2. Let “ p ” denote the probability with which Inspector plays the pure strategy “Inspect” (so “1-p” is the probability with which he plays “don’t inspect”) , and let “ q ” denote the probability with which the Thief plays “Rob” (so “1-q” is the probability with which he plays “don’t rob”) . Notice: Play (inspect) with probability p, play (don’t inspect) with probability 1-p “ IS A MIXED STRATEGY for the Inspector. 2
Play (Rob) with probability q, play (don’t rob) with probability 1-q “ IS A MIXED STRATEGY for the Thief. The Inspector ’s expected payoff under these mixed strategies is: E U Inspector = (probability that “inspect,rob” is played). 10 + (probability that “inspect,don’t” is played). 0 + (probability that “don’t inspect,rob” is played). 0 + (probability that “don’t inspect, don’t rob” is played). 5 = = p [q].10 + [ p .(1-q)].0 + [ (1-p) q].0 + [ (1-p) (1-q)].5 = 10 p q + (1-p) [5(1-q)] The Thief ’s expected payoff under these mixed strategies is: EU Thief = (probability that “inspect,rob” is played). 0 + (probability that “inspect,don’t” is played).

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

View Full Document
This is the end of the preview. Sign up to access the rest of the document.

## This note was uploaded on 03/16/2012 for the course FENS 101 taught by Professor Selçukerdem during the Fall '12 term at Sabancı University.

### Page1 / 19

06.MixedStrategy - MIXED STRATEGIES A PURE STRATEGY is...

This preview shows document pages 1 - 4. Sign up to view the full document.

View Full Document
Ask a homework question - tutors are online