Simulations accompanying Mean Field Equilibria for Multiarmed Bandit Games

This repo contains the Python code used to run simulations for the Mean Field Equilibria for Multiarmed Bandit Games paper. A preprint can be found on SSRN.

Requirements

• Python 3
• Jupyter notebooks
• MCSim: small framework for Monte Carlo simulations
• Toyplot: library used to create plots
• Statsmodels: statistics library used to fit LOWESS

Notebooks

High level code to run simulations and plot results are Jupyter notebooks:

• simulation_playground: this notebook gives a simple example of how to set up a particular simulation, and useful if you want to simulate a specific scenario
• run_sims_disk: this notebooks runs the simulations found in the paper and saves the outcomes as pickle objects. Before using this notebook, run setup.sh to create subdirectories used to store simulations and plots.
• plots: this notebook recreates the plots used in the paper. It reads the pickle files created by the run_sims_disk notebook, so run that first.

Questions

If you have any questions about the code or simulations, or have trouble getting things to work, please open an issue.