Simulation and approximation code for the paper "Local Symmetry and Global Structure in Adaptive Voter Models" by Phil Chodrow and Peter Mucha.
The code for simulating adaptive voter models was lightly modified from that provided by Feng Bill Shi, available here.
The simulation code is implemented in C++. While it is possible to call the simulation directly, a better solution for bulk computation is to interact with the C++ code via the file C/run.py, which handles multiple simulations with varying parameters.
Requirements
- Python 2.7+
- Python modules:
numpy,subprocess.
The novel approximation method described in the paper is implemented in R.
While these functions may be called directly, an easier approach is to program parameter ranges into compute_arch.R.
Requirements
R3.3+Rpackages:tidyverse,alabama
The file figs.rmd is sufficient to produce all figures in the paper, with the exception of Figure 1, once both C/run.py and compute_arch.R have been run with their provided parameter values. Note that multiple days of computation time may be required to produce the complete set of results and figures from scratch.
Requirements
R3.3+Rpackages:tidyverse,ggjoy,data.table,RColorBrewer