collpcm: Bayesian model selection in latent position cluster models for networks

The latent position cluster model is a popular model for the statistical analysis of network data. This model attaches a latent position to vertices in the network graph, with the motivation that the distance between two latent positions is connected to the probability of there being an edge between the corresponding vertices in the graph. Bayesian methods are used to estimate the likely configurations of the latent positions based on a prior assumption on their layout. This prior assumption can reflect a clustering or grouping in the latent positions which can be used to say something about community formation between vertices in the network graph. This package provides convenient computational facilities to fit latent position cluster models, while exploring probable community patterns in the network i.e. exploring the likely groupings or clustering present. The methods do not require specifying a number of groupings or communities in the network at the outset.