lecture7 notes

# 10 game theory lecture 7 uniqueness of a pure strategy

This preview shows page 1. Sign up to view the full content.

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

Unformatted text preview: ( ¯ − x ∗ )T �u (x ∗ ) + (x ∗ − x )T �u (x ) x ¯ ¯ (10) ∗ ∗T ∗ T∗ ¯ + ∑ λ �hi (x ) (xi − x ) + ∑ λi �hi (xi ) (x − xi ) ¯ ¯ ¯ i i i i ∈I > i i ∈I ¯ ¯ ¯ ¯ ∑ λi∗ �hi (xi∗ )T (xi − xi∗ ) + ∑ λi �hi (xi )T (xi∗ − xi ), i ∈I i ∈I where to get the strict inequality, we used the assumption that the payoﬀ functions are diagonally strictly concave for x ∈ S . Since the hi are concave functions, we have hi (xi∗ ) + �hi (xi∗ )T (xi − xi∗ ) ≥ hi (xi ). ¯ ¯ 10 Game Theory: Lecture 7 Uniqueness of a Pure Strategy Equilibrium Proof Using the preceding together with λi∗ > 0, we obtain for all i , λi∗ �hi (xi∗ )T (xi − xi∗ ) ¯ ≥ λi∗ (hi (xi ) − hi (xi∗ )) ¯ = λi∗ hi (xi ) ¯ ≥ 0, where to get the equality we used Eq. (7), and to get the last inequality, we used the facts λi∗ > 0 and hi (xi ) ≥ 0. ¯ Similarly, we have ¯ λi �hi (xi )T (xi∗ − xi ) ≥ 0. ¯ ¯ Combining the preceding two relations with the relation in (10) yields a contradiction, thus concluding the proof. 11 Game Theory: Lecture 7 Uniqueness of a Pure Strategy Equilibrium Suﬃcient Condition for Diagonal Strict Concavity Let U (x ) denote the Jacobian of �u (x ) [see Eq. (5)]. In particular, if the xi are all 1-dimensional, then U (x ) is given by ⎛ ∂2 u (x ) ∂2 u (x ) ⎞ 1 1 ··· 2 ∂x1 ∂x2 ⎜ ∂x1 ⎟ ⎜2 ⎟ .. U (x ) = ⎜ ∂ u2 (x ) ⎟. . ⎝ ∂x2 ∂x1 ⎠ . . . Proposition For all i ∈ I , assume that the strategy sets Si are given by Eq. (3), where hi is a concave function. Assume that the symmetric matrix (U (x ) + U T (x )) is negative deﬁnite for all x ∈ S , i.e., for all x ∈ S , we have y T (U (x ) + U T (x ))y < 0, ∀ y �= 0. Then, the payoﬀ functions (u1 , . . . , uI ) are diagonally strictly concave for x ∈ S . 12 Game Theory: Lecture 7 Uniqueness of a Pure Strategy Equilibrium Proof Let x ∗ , x ∈ S . Consider the vector ¯ x ( λ ) = λx ∗ + (1 − λ )x , ¯ for some λ ∈ [0, 1]. Since S is a convex set, x (λ) ∈ S . Because U (x ) is the Jacobian of �u (x ), we have d �u (x (λ)) dλ = or �1 0 dx (λ) d (λ) U (x (λ))(x ∗ − x ), ¯ = U (x (λ)) U (x (λ))(x ∗ − x )d λ = �u (x ∗ ) − �u (x ). ¯ ¯ 13 Game Theory: Lecture 7 Uniqueness of a Pure Strategy Equilibrium Proof Multiplying the preceding by (x − x ∗ )T yields ¯ ( ¯ − x ∗ )T �u (x ∗ ) + (x ∗ − x )T �u (x ) x ¯ ¯ �1 1 =− (x ∗ − x )T [U (x (λ)) + U T (x (λ))](x ∗ − x )d λ ¯ ¯ 20 > 0, where to get the strict inequality we used the assumption that the symmetric matrix (U (x ) + U T (x )) is negative deﬁnite for all x ∈ S . 14 Game Theory: Lecture 7 Supermodular Games Supermodular Games Supermodular games are those characterized by strategic complementarities Informally, this means that the marginal utility of increasing a player’s strategy raises with increases in the other players’ strategies. Implication ⇒ best response of a player is a nondecreasing f...
View Full Document

## This document was uploaded on 03/19/2014 for the course EECS 6.254 at MIT.

Ask a homework question - tutors are online