This README describes how to replicate the analysis in our manuscript.

Part A: Simulation of Infectious Process on Contact Networks.

1. Use R to run simulation.R For values 1-2000 of SEED. (The code is arranged this way for easy paralellization.) Note that this code calls data_code.py, which generates the synthetic contact networks and performs the epidemic process.
2. Use R to run analysis.R for values 1-2000 of SEED. This code is also designed for paralellization. This creates all relevant simulation output. Note that this will require downloading the OBSgeeDR package for observational augmented GEE analysis.
3. Use R to run "global analysis.R" for final formatting.

Part B: Application to Microfinance Data.

1. The microfinance data used in this paper is publically available, referenced here: http://economics.mit.edu/faculty/eduflo/social
2. use Python to run data generation.py to compile the microfinance data into a single dataset.
3. Use R to run Data Completion.R to concatenate the data with the defined exposures.
4. Use R to run Empirical Data Analysis.R to create the tables in the paper and supplement.

For any questions, please contact the corresponding author (Patrick Staples) at patrickstaples at fas dot harvard dot edu.