Title: A semiparametric extension of the stochastic block model for longitudinal networks
Authors: Tabea Rebafka - Pierre et Marie Curie University (France) [presenting]
Catherine Matias - CNRS - Universite Pierre et Marie Curie (France)
Fanny Villers - Pierre et Marie Curie University (France)
Abstract: To model recurrent interaction events in continuous time, an extension of the stochastic block model is proposed where each individual belongs to a latent group and interactions between two individuals follow a conditional inhomogeneous Poisson process whose intensity is driven by the individuals' latent groups. The model is shown to be identifiable and an estimation procedure is proposed 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, the performance of the new procedure is demonstrated on synthetic experiments and real datasets are analysed to illustrate the utility of the approach.