A semiparametric extension of the stochastic block model for longitudinal networks

TL;DR

This paper introduces a semiparametric extension of the stochastic block model for analyzing longitudinal networks with recurrent interactions, employing a variational EM algorithm and nonparametric intensity estimation.

Contribution

It presents a novel semiparametric model for longitudinal networks with latent groups and develops an estimation method using variational EM with nonparametric intensity estimators.

Findings

Model is identifiable and effectively estimates latent groups.

Method performs well on synthetic data and real datasets.

Adaptive nonparametric intensity estimation improves model flexibility.

Abstract

To model recurrent interaction events in continuous time, an extension of the stochastic block model is proposed where every individual belongs to a latent group and interactions between two individuals follow a conditional inhomogeneous Poisson process with intensity driven by the individuals' latent groups. The model is shown to be identifiable and its estimation is based on a semiparametric variational expectation-maximization algorithm. Two versions of the method are developed, using either a nonparametric histogram approach (with an adaptive choice of the partition size) or kernel intensity estimators. The number of latent groups can be selected by an integrated classification likelihood criterion. Finally, we demonstrate the performance of our procedure on synthetic experiments, analyse two datasets to illustrate the utility of our approach and comment on competing methods.