Scribd
Upload
36 views

Normal Form Games

Uploaded by

Kavisha Patel
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
36 views

Normal Form Games

Uploaded by

Kavisha Patel
Copyright
© © All Rights Reserved
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 6

Game theory I : Normal Form Games

Objective of Game Theory:


- Domain /Objective:
- Situations in which two or more players strategically interact
- Offer guidance as to how such situations can be resolved

Definition of a game: what constitutes a game?


- Two or more player
- Set of possible actions (prices)
- Payoffs (profits,rewards,utility)
- Plus certain additional info on game structures such as:
o Ability to coordinate
o Timing of moves
o How often each player moves
o Info the players have

Types of games:
Use additional info to classify a game into
- Non–cooperative/cooperative
- Simultaneous/sequential game
- One shot/repeated game
- Perfect/imperfect information game
Here: simultaneous move, one-shot game with perfect information (non–cooperative); also
called normal form games

Representation of normal form games:


Simplest form: Payoff Matrix
Game theory... gives us a prediction about how this situation might be resolved if players A
and B are rational, have perfect information and play only once.

Example:

Strategies: plans over (future) actions.

Pure strategies si
are simple plans to pick one particular action, e.g. I will charge £50 for my product

Mixed strategies σi
are randomisations over pure strategies, e.g.,
I will charge a price of £50 with 75 percent probability and a price of £40 with 25 percent
probability

Conditional strategies
are plans that depend on the realisation of something observable, e.g.,
I will charge a price of £50 if my competitor did the same in the last period; otherwise, I will
charge a price of £40.
This is an example of a trigger-price strategy. Famous examples are:
Carrot-and-Stick strategy and
Tit-for-tat strategy

Example:
A pure strategy – you have a plan to pick one particular option and so mewQ = (1,0)

Solution concepts:
- Dominance
- Iterated dominance
- Nash equilibrium
- Refinements of the Nash Equilibrium

Dominant strategies:

A dominant strategy is a strategy that gives a player a pay- off that is at least as large as that
of all the other strategies, no matter what strategy is chosen by the opponent
Dominated strategy:

A strategy is dominated if there is another strategy that gives the player payoffs that are at
least as high as the pay- offs for the dominated strategy, no matter what strategies are chosen
by the opponents.

Dominant strategy vs dominated strategy:


Dominant Strategy A dominant strategy is the best strategy for a player to follow regardless
of what the other players do. In other words, choosing this strategy yields the highest payoff
for a player, no matter what strategies the opponents choose. If a player has a dominant
strategy, they can make their decision without worrying about the actions of others because
their best choice is clear and unchanging.

Example: In a prisoner's dilemma game, if betraying your accomplice gives you a better
outcome (lesser jail time) regardless of whether your accomplice betrays you or remains
silent, then betraying is a dominant strategy.

Dominated Strategy A dominated strategy, on the other hand, is a strategy that results in
the worst outcome for a player, no matter what the other players do. If there exists another
strategy that always produces a better outcome than the dominated strategy, regardless of
the opponents' actions, the inferior strategy is considered dominated. Rational players will
never choose a dominated strategy because there is always a better alternative available.

Example: If remaining silent always results in more jail time compared to betraying
(regardless of the accomplice's choice), then remaining silent is a dominated strategy.

Equilibrium in dominant strategies:


A game has an equilibrium in dominant strategies if both/all players have a dominant strategy.
BOTH PLAYERS HAVE A DOM STRAT = EQUILIBRIUM CAN BE REACHED IN A GAME

Notes:

- No rational player would choose a (strictly) dominated strategy

- (Strictly) dominated strategies can be disregarded

- Each player can have at most one dominant strategy

- Equilibrium in dominant strategies is always unique

Example 1:
Player A:

Player B:

Equilibrium prediction:

Example 2:

Iterated dominance:
- Iterated Dominance means that we repeatedly delete strictly dominated strategies.
- If only one combination of pure strategies remains, this will be the obvious prediction
(equilibrium).
- Outcome of (strictly) iterated dominance does not depend on the order.
- It doesn’t matter whether we start with player A or B.
- Unfortunately, the same is not true for weak dominance

Weak dominance:
- A strategy is weakly dominated by another strategy if the other strategy yields a
strictly higher payoff for at least one strategy of the opponent and the same payoffs for
all other strategies of the opponent.

Motivation:
- There are many situations in which players do not have a dominant strategy and where
iterated dominance doesn’t help

Nash Equilibrium….. is a strategy profile such that the strategy maximises the player’s payoff
given the other player plays the equi- librium strategy,
i.e. each player’s strategy is a best response to the other player’s strategy.

Strategy profile:

Example 3:

Example 4:

You might also like