An R package for simulating dynamics on networks. The primary motivations for this package were: (1) simulating dynamical equations coupled through an adjacency matrix (or Laplacian matrix, but I haven't tried it yet), and (2) the need to switch back and forth between deterministic and stochastic simulations of the same dynamics. For ODEs, relies on the deSolve package. For SDEs, implements some of the ideas in the sde package and its accompanying book with the goal of interoperability with deSolve.
Version 0.2-1 introduces a breaking change: the default models, which I used in several projects, have been switched from adjacency matrix to adjacency list computation. The new version should be much faster.
Recently updated to include models from Barzon et al. 2024, but check parameter values and ranges to confirm desired behavior. Code for new models is in ./R/barzon-dynamics.R. Other models can be found in ./R/models.R and ./R/normal-forms.R.
For now, I recommend cloning this repository, then using the following shell commands to build and install the package (assuming your working directory is the one that contains the cloned sdn directory):
R CMD build sdn
R CMD INSTALL sdn_0.2-1.tar.gz