Bergm: Bayesian exponential random graph models in R

TL;DR

Bergm is an R package that enables Bayesian inference for network data using MCMC, providing tools for model fitting, goodness-of-fit assessment, and uncertainty quantification, suitable for large networks.

Contribution

It introduces a user-friendly R package that extends ergm with Bayesian inference capabilities and improved speed for large network analysis.

Findings

Enhanced speed performance for large networks

Provides Bayesian goodness-of-fit procedures

Facilitates probabilistic uncertainty analysis

Abstract

The Bergm package provides a comprehensive framework for Bayesian inference using Markov chain Monte Carlo (MCMC) algorithms. It can also supply graphical Bayesian goodness-of-fit procedures that address the issue of model adequacy. The package is simple to use and represents an attractive way of analysing network data as it offers the advantage of a complete probabilistic treatment of uncertainty. Bergm is based on the ergm package and therefore it makes use of the same model set-up and network simulation algorithms. The Bergm package has been continually improved in terms of speed performance over the last years and now offers the end-user a feasible option for carrying out Bayesian inference for networks with several thousands of nodes.