No abstract

The Exponential-family Random Graph Model (ERGM) is a powerful statistical model to represent the complicated structural dependencies of a binary network observed at a single time point. However, regarding dynamic valued networks whose observations are matrices of counts that evolve over time, the development of the ERGM framework is still in its infancy. We propose a Partially Separable Temporal ERGM (PST ERGM) for dynamic valued networks to facilitate the modeling of dyad value augmentation and dyad value diminution. Our parameter learning algorithms inherit state-of-the-art estimation techniques to approximate the maximum likelihood, by drawing Markov chain Monte Carlo (MCMC) samples conditioning on the network from the previous time step. We demonstrate the ability of the proposed model to interpret network dynamics and forecast temporal trends with real data.

No abstract

No abstract