Problem Set
Game Theory
Tutorial session February 21, 2025
1. A police officer and a robber are engaged in a strategic battle of wits. The police
officer (Player 1) must decide whether to patrol the city in search of criminals or
to stay at the police station and rest. His payoff from staying at the station is 1,
as he enjoys a relaxed shift. However, if he chooses to patrol, his reward depends
on whether he catches the robber. The robber (Player 2) must decide between
committing a crime or lying low. If the robber remains inactive and avoids
committing any crimes, his payoff is 0, as he neither gains nor loses anything; in
this case the payoff for the police is also 0. However, if he attempts a robbery,
the outcome depends on whether the police officer is patrolling. If the officer is
out on patrol, he will catch the robber, resulting in a payoff of 4 for the officer
and −2 for the robber (who gets arrested). But if the officer chooses to stay
at the station, the robber successfully commits the crime, gaining a payoff of 2,
while the officer, unaware of the crime, still gets 1 from his relaxed shift.
Find the mixed strategy nash equilibria in this game.
2. The game table:
P 1\P 2 Lef t Center Right
T op (1.5, 2.5) (0, 4) (0, 3)
M iddle (2, 0) (0.5, 1.5) (−1, 3)
Bottom (1, 0) (1, −1) (−0.5, 0.5)
Find all the mixed strategy Nash Equilibriums in the above game table
through IESDS.
Note: Definition of Iterated Elimination of Strictly Dominated Strategies(IESDS):-
The process of eliminating repeatedly any strictly dominated strategy or strategies
(i.e., a strategy that always yields a lower payoff compared to another).
3.
Definition 1. In a game, player i’s mixed strategy αi strictly dominates her
action a′i if
Ui (αi , a−i ) > Ui (a′i , a−i ) ∀a−i
Theorem 1. The action a′i , which is strictly dominated by a mixed strategy αi ,
is played with zero probability at the mixed strategy nash equilibria.
Consider the following game:
Player 2
A B
P (0, 1) (4, 0)
Player 1
Q (1, 2) (1, 4)
R (2, 0) (0, 1)
(i) Show that action Q is not strictly dominated by pure strategies P and R.
1 2
(ii) Show that the mixed strategy p(P ) = 3 , p(Q) = 0 p(R) = 3 , strictly
dominates Q.
(iii) Using the theorem described above, find the mixed strategy nash equilibria
of this game.
4. Consider the following game,
P 1\P 2 P lay P ass
P lay (1, 2) (3, 3)
P ass (2, 2) (0, 2)
(i) Write the players’ total expected payoffs.
(ii) Draw the best response function.
(iii) Write down all the mixed strategy Nash equilibriums.
5. (Dividing a cake fairly) Two players use the following procedure to divide a cake.
Player 1 divides the cake into two pieces, and then Player 2 chooses one of the
pieces; Player 1 obtains the remaining piece. The cake is continuously divisible
(no lumps!), and each player likes all parts of it.
Page 2
(i) Suppose that the cake is perfectly homogeneous, so that each player cares
only about the size of the piece of cake she obtains. How is the cake divided
in a subgame perfect equilibrium?
(ii) Suppose that the cake is not homogeneous: the players evaluate different
parts of it differently. Represent the cake by the set C, so that a piece of
the cake is a subset P of C. Assume that if P is a subset of P’ not equal
to P’ (smaller than P’) then each player prefers P’ to P. Assume also that
the players’ preferences are continuous: if player i prefers P to P’ then
there is a subset of P not equal to P that player i also prefers to P’. Let
(P1 , P2 ) (where P1 and P2 together constitute the whole cake C) be the
division chosen by player 1 in a subgame perfect equilibrium of the divide-
and-choose game, P2 being the piece chosen by player 2. Show that player
2 is indifferent between P1 and P2 , and player 1 likes P1 at least as much
as P2 . Give an example in which player 1 prefers P1 to P2 .
Page 3
