Edit on GitHub

GameTheory.jl

GameTheory.jl is a Julia package about algorithms and data structures for Game Theory.

Installation

To install the package, enter the Pkg mode by pressing ] and run

add GameTheory

Usage

Once installed, the GameTheory package can be used by typing

using GameTheory

The Base type Player can be created by passing a payoff matrix:

player1 = Player([3 1; 0 2])
2×2 Player{2, Int64}:
 3  1
 0  2

A 2-player NormalFormGame can be created either by passing Player instances,

player2 = Player([2 0; 1 3])
g = NormalFormGame((player1, player2))
print(g)
2×2 NormalFormGame{2, Int64}:
 [3, 2]  [1, 1]
 [0, 0]  [2, 3]

or by passing a payoff matrix directly:

payoff_bimatrix = Array{Int}(undef, 2, 2, 2)
payoff_bimatrix[1, 1, :] = [3, 2]
payoff_bimatrix[1, 2, :] = [1, 1]
payoff_bimatrix[2, 1, :] = [0, 0]
payoff_bimatrix[2, 2, :] = [2, 3]
g = NormalFormGame(payoff_bimatrix)
print(g)
2×2 NormalFormGame{2, Int64}:
 [3, 2]  [1, 1]
 [0, 0]  [2, 3]

After constructing a NormalFormGame, we can find its Nash Equilibria by using methods of GameTheory. For example, pure_nash finds all pure action Nash Equilibria by enumeration:

pure_nash(g)
2-element Vector{Tuple{Int64, Int64}}:
 (1, 1)
 (2, 2)

Please see the notebooks on QuantEcon for more details.

Notebooks

Some notebooks for demonstration are available:

Library Outline