• 検索結果がありません。

最近の更新履歴 yyasuda's website

N/A
N/A
Protected

Academic year: 2017

シェア "最近の更新履歴 yyasuda's website"

Copied!
2
0
0

読み込み中.... (全文を見る)

全文

(1)

Solution Keys to the Midterm Exam (Game Theory, Winter 2008)

1. Nash Equilibrium (10 points)

A strategy pro…le s is a Nash equilibrium if

ui(s ) ui(s i; si) for all i and si. 2. True or False (15 points)

(a) False (b) True (c) True

3. Nash equilibrium and dominated strategies (25 points) (a) (B; Z) is the unique Nash equilibrium.

(b) We will get (B; Z) in the following iterated elimination process: Step 1: We can erase X since X is strictly dominated by Z.

Step 2: Given step 1, we can erase A since A is strictly dominated by B.

Step 3: Given steps 1 and 2, we can erase Y since Y is strictly dominated by Z. (c) Any combinations of x and y that satisfy x + y = 100 are Nash equilibria. Clearly, there are 101 such equilibria, i.e., (0; 100)(1; 99):::(100; 0).

4. Mixed strategy equilibrium (25 points)

(a) Given that your opponent chooses actions completely randomly, your expected payo¤ becomes 0 no matter how you play. Therefore, choosing each action with prob- ability 1=3 is indeed a (weakly) best response. Thus, when both players choose actions completely randomly, their strategies are mutually best response to each other, and hence constitute a Nash equilibrium.

(b) Let p be a probability that player 2 would choose Rock, and q be a probability that she chooses Paper. Note that her probability of choosing Scissors is written as 1 p q. Under mixed strategy Nash equilibrium, player 1 must be indi¤erent amongst choosing Rock, Paper and Scissors, which implies that these three actions must give him the same expected payo¤s. Let uR; uP; uS be his expected payo¤s by selecting Rock, Paper and Scissor respectively. Then, they can be calculated as follows:

uR = q + 2f1 p qg uP = p f1 p qg uS = 2p + q

1

(2)

Since uR= uP, we obtain

q + 2f1 p qg = p f1 p qg

, 3f1 (p + q)g = p + q

, p + q = 3=4.

Since uP = uS, we obtain

p f1 p qg = 2p + q , 4p = 1

, p = 1=4.

Substituting into p+q = 3=4, we achieve q = 1=2. Since the game is symmetric, we can derive exactly the same result for Player 1’s mixed action as well. Therefore, we get the mixed-strategy Nash equilibrium: both players choose Rock, Paper and Scissors with probabilities 1=4; 1=2; 1=4 respectively.

(c) There are two pure-strategy Nash equilibria: (A; X) and (B; Y ).

(d) Let p be a probability that player 2 chooses X and q be a probability that player 1 chooses A. Since player 1 must be indi¤erent amongst choosing A and B, we obtain

2p = p + 3(1 p) , 4p = 3 , p = 3=4.

Similarly, player 2 must be indi¤erent amongst choosing X and Y , which implies 4q + 6(1 q) = 7(1 q)

, 5q = 1 , q = 1=5.

Thus, the mixed-strategy equilibirum is that player 1 takes A with probability 1=5 (and B with probability 4=5) and player 2 takes X with probability 3=4 (and Y with probability 1=4).

5. Focal point (5 points)

There is no …xed answer of this question. But, you would be likely receive 5 points if you chose “Game Theory.”

2

参照

関連したドキュメント

To complete the “concrete” proof of the “al- gebraic implies automatic” direction of Theorem 4.1.3, we must explain why the field of p-quasi-automatic series is closed

In general, Liouville type theorems for stable solutions of nonlinear elliptic equations are usually guaranteed in low dimensional case.. The main purpose of this paper is to obtain

[14.] It must, however, be remembered, as a part of such development, that although, when this condition (232) or (235) or (236) is satisfied, the three auxiliary problems above

These estimates involve three exponents which are usually associated to infinite graphs: the spectral dimension δ s (which is by definition twice the exponent of n − 1 in local

A crucial physical prescription is that the field must be covariant under the action of a unitary representation U(g) of some transformation group (such as the Poincaré or Lorentz

We prove that any simply connected and complete Riemannian manifold, on which a Randers metric of positive constant flag curvature exists, must be diffeomorphic to an

Kartsatos, The existence of bounded solutions on the real line of perturbed non- linear evolution equations in general Banach spaces, Nonlinear Anal.. Kreulich, Eberlein weak

In fact, we have shown that, for the more natural and general condition of initial-data, any 2 × 2 totally degenerated system of conservation laws, which the characteristics speeds