Solutions to Homework #3

0.1

3.42

By Theorem 3.5, we can recover the production function by solving the maximization problem
f (x1 , x2 ) = max y

s.t. w1 x1 + w2 x2 = c(w1 , w2 , y)

w1 ,w2

Suppose c(w1 , w2 , y) = yAw1↵ w21

. Then the problem is
s.t. w1 x1 + w2 x2 = yAw1↵ w21

f (x1 , x2 ) = max y
w1 ,w2

= max

w1 ,w2

w 1 x1 + w 2 x2
Aw1↵ w21 ↵

The first-order conditions are
@
x1
=

@w1
Aw1 w21
@
x2
=

@w1
Aw1 w21

↵(w1 x1 + w2 x2 )
=0
w1 Aw1↵ w21 ↵

(1

↵)(w1 x1 + w2 x2 )
=0
w2 Aw1↵ w21 ↵

Combining these two, we get

w 1 x1
w 2 x2
=

1 ↵
which is the same first-order conditions as the Cobb-Douglas production function f (x1 , x2 ) =
x↵1 x12 ↵ .
1

Suppose c(w1 , w2 , y) = yA(w1r + w2r ) r . Then the problem is
1

s.t. w1 x1 + w2 x2 = yA(w1r + w2r ) r

f (x1 , x2 ) = max y
w1 ,w2

= max

w1 ,w2

w 1 x1 + w 2 x2
1

A(w1r + w2r ) r

The first-order conditions are:
@
x1
=
1
@w1
A(wr + wr ) r

w 1 x1 + w 2 x2
1
r 1
w1 A(w1r + w2r ) r +1

=0

@
x2
=
1
r
@w2
A(w1 + w2r ) r

w 1 x1 + w 2 x2
1
r 1
w2 A(w1r + w2r ) r +1

=0

1

Combining these two, we get

2

x2
x1
r 1 =
w2
w1r 1

which is the same first-order conditions as the CES production function.
1

the problem becomes 25 min wk k + wF k k with first-order condition @ 25 = wk wF 2 = 0 @k k ⇣ ⌘1 2 The solution is k = 5 wwFk . 0. Let A be a strictly dominated strategy. M }. Now. -1 (a) In this game. then M . and every round of elimination must remove at least one strategy.t. then R. There are no further weakly dominated strategies. (b) First.1 3. iterative elimination of strictly dominated strategies (IESDS) terminates in a finite number of rounds. A will always be eliminated by IESDS. Suppose instead.2 3.-2 M 1. (kF1 ) 2 = 4. Therefore. it will remain strictly dominated for the rest of IESDS until it is eliminated. Then. 2 . (kF2 ) 2 = 3 k.F2 Rearranging the constraints.-1 R 0. C. Since there are a finite number of strategies.1 1. F2 = 9/k. any subset of A must also be strictly dominated. Suppose we eliminate D.0. C} and L.F1 . any strategy that becomes strictly dominated in the process of IESDS will eventually be eliminated. A remains strictly dominated.3 U C D L 2.3 7.1 1. D} and {L. we will prove 7. no matter what the order of elimination is. we eliminate R. U weakly dominates D and M weakly dominates R. We are left with the 2 ⇥ 1 game consisting of {U. then IESDS will eventually eliminate A in a finite number of rounds. and that once a strategy becomes strictly dominated.0 2.2 2. that is. and the process of elimination must terminate if each player has only 1 strategy left. we get F1 = 16/k. no matter what other strategies are eliminated. We have established that all strategies that are strictly dominated at the beginning of IESDS will eventually be eliminated. Therefore. suppose we have a finite game with at least one strictly dominated strategy. so what remains is the 3 ⇥ 2 game consisting of {U. then IESDS terminates in a finite number of rounds. Since IESDS terminates in a finite number of rounds.1 -1.2: For a finite game. Plugging this into the objective function.55 The problem is to minimize total cost: 1 min wk k + wF (F1 + F2 ) 1 s.

1] is a best response. E2 (R) = 2p 3 1 . . Player 1’s best response is p = 1. 1] is a best response of Player 2).6 Suppose we have a sequence of eliminated. Likewise.an . all p 2 [0. when p > 0. Suppose we have an strategy A such that there exists a sequence of eliminations of strictly dominated strategies a1 . R). If E1 (U ) = E1 (D). Ak+1 . (U.. (U. let A0 be the set of strategies that are strictly dominated at the beginning of IESDS. Then. Ak 1 no matter what the order of elimination is.. Let’s find the set of mixed strategy NE. 0. .Finally. .. E1 (D) = 0 E2 (L) = p. when q > 0. then there does not exist a sequence of eliminations of strictly dominated strategies that eliminates A. Player 2’s best response is q = 1. Since a strictly dominated strategy is also a weakly dominated strategy.. Taking the contrapositive... L).4 7.. L) has neither player playing a weakly dominated strategy. (b) There are three pure NE. there is some k > 0 for which Ak . R). A1 be the set of remaining strategies that are strictly dominated after all strategies in A0 have been eliminated.an . Since IESDS terminates in a finite number of rounds. L). 0. Let p be Player 1’s probability of playing U . a2 . IESDS will eventually eliminate all strategies in A0 . (U.. E2 (R) = 0 If E1 (U ) < E1 (D). which are the two pure NE we have already found. (U.5 7. this is also a sequence of eliminations of weakly dominated strategies. L) and (D. 1] is a best response of Player 1). E1 (U ) = E1 (D) when q = 0 (therefore all p 2 [0. E1 (D) = 2q 1 E2 (L) = p. L) is one where neither player plays a weakly dominated strategy. A1 . etc. . The expected payo↵s to pure strategies of each player are: E1 (U ) = q. then Player 1’s best response is p = 0. are empty. The expected payo↵s to pure strategies of each player are: E1 (U ) = q. and (D. that eliminates A. if A is such that there does not exist a sequence of eliminations of weakly dominated strategies that eliminates A. and q be Player 2’s probability of playing R. there exists a sequence of eliminations of weakly dominated strategies that eliminates A (the same sequence)..10 (a) There are two pure NE. and q be Player 2’s probability of playing R. If E1 (U ) > E1 (D). Only (U. strictly dominated strategies a1 . The only intersections are p = q = 0 and p = q = 1. E2 (L) = E2 (R) when p = 0 (therefore all q 2 [0. A2 be the set of remaining strategies that are strictly dominated after all strategies in A0 and A1 have been eliminated. Let p be Player 1’s probability of playing U . Player 1’s best response is p = 1.

and (D. 2. Now. this relation- Suppose all three players play the same mixed strategy. which is .e. Then 1 = satisfied when one of p. The set of intersections is : p = 1. r. P ) is not a NE. 1). • If one player chooses D. since any player can increase his payo↵ from to D. P. since the D player will decrease his payo↵ from 0 to 1 by switching. Let p. 1] . 3). m). Suppose one player (say. r to be a NE. L). The pure NE are (U. it is a NE. the payo↵ vector is ( 3. Player 1’s best response is p = 1. play P . Any completely mixed strategy (i. E2 (L) = E2 (R) when p = 1 (therefore all q 2 [0. e. 0. r be the probability that Players 1. Player 2’s best response is q = 1.g. D) or (D. l).g. 1] is a best response of Player 1).6 7. e. when q < 1. Player 2) plays pure P . • (D. We take the position of Player 1. (c) M is strictly dominated by L and can be eliminated. P ). 1. e. D. P. 1] is a best response of Player 2). 4 3r 1 6r 3 and q = 3 1 6 3. q. • If two players choose D. e. The probability that the other two players play (P. (D. Then r = giving p = q = r = 0. P. when p < 1. q = 1. q)(1 q)(1 r)( 4) r)( 3) In order for p. (D. D. (D. P ) is (q(1 r) + (1 q)r). since any player will decrease his payo↵ from 3 to 4 by switching. we find the mixed NE. q. the only NE where no player plays a weakly dominated strategy is (U. q)r)( 1) + (1 q)r)( 3) + (1 3r 1 6r 3 . and the P player will decrease his payo↵ from 1 to 3 by switching. 3r 1 6r 3 . P ). we find the pure strategy NE.g. D) is (1 q)(1 r). • If three players choose D. P ) is qr. D. 1). P ). 3. • (P. since a D player can increase his payo↵ from 3 to 1 by switching.2113. The probability that the other players play (D. (D. L). 1 to 0 by switching • When 1 player chooses D. the payo↵ vector is (0.E1 (U ) = E1 (D) when q = 1 (therefore all p 2 [0. 1. D) is a NE. q. q 2 [0. Therefore. (C.g. which we have already found.14 • If all players choose P : (P. The probability that the other players play (P. the payo↵ vector is ( 3. D). First. • When 2 players choose D. The expected payo↵ to pure strategies is: E1 (P ) = qr( 1) + (q(1 E1 (D) = qr(0) + (q(1 r) + (1 r) + (1 Equating these two gives the condition q = ship must hold between any pair of p. (D. 4). P ). 3. 1] and p 2 [0. Likewise. q. then p = q = r. it is not a NE. P. not a pure strategy) is weakly dominated by the pure strategy U (for Player 1) or L (for Player 2). P ). the payo↵ vector is ( 1. 3 respectively. r is 1 and the other two are 32 . D.

Let q be the probability that Player 2 plays S. Spy which uses a weakly dominant strategy. Player 2’s expected payo↵ to pure strategies are: E2 (S) = ✏( 9) + (1 E2 (D) = ✏( 1) + (1 E2 (S) = E2 (D) when p = ✏)(p + 2(1 ✏)( p + 2(1 p)) p)) 4✏ 1 ✏. 0. 5 .8 7. Keep is strictly dominant. (b) Spy is weakly dominant. Since a mixed strategy is never weakly dominant. and Player 2 plays a mixed strategy of (0. The other NE is where Player 1 plays Destroy. (a) For the ”aggressive” type. so it will always be played. 0. q = 12 . Therefore.0. this is the NE where no player plays a weakly dominant strategy. (c) The only pure strategy N E is Destroy. Let us call the ”non-agressive” type as Player 3.17 (a) Neither player has a strictly dominant strategy. (b) Let p be the probability that the ”non-aggressive” type of Player 1 plays Keep.7 7. Player 3’s expected payo↵s to pure strategies are: E3 (Keep) = q + (1 q) = 1 2q E3 (Destroy) = 0 E3 (Keep) = E3 (Destroy) when q = 12 . a Bayesian-Nash equilibrium is when p = 14✏✏ .18 We will denote Spy as S and Don’t Spy as D.5).5.