An Agent Based Artificial Market

Running the Model

To run and interact with the model, run analysis.ipynb through Jupyter Notebook or a similiar Notebook viewer.

All logic is contained in market.py, agent.py and model.py, with model.py providing a method for initializing the model, view the notebook for example usage.

Usage

from model import MarketModel

# Parameterise the model (View the notebook for example parameters)

# Initialize the model
model = MarketModel(steps=timesteps, agents=agents, non_random=non_random, 
                    price=price, alpha=alpha,
                    beta=beta, gamma=gamma, rho=rho, 
                    network=network, convergence=convergence)

# Run the model
model.run()

Dependencies (Python Packages)

• mesa
• networkx
• statsmodels
• numpy, pandas, matplotlib