Dynamic latent space relational event model

Dynamic relational processes, such as e-mail exchanges, bank loans and scientific citations, are important examples of dynamic networks, in which the events consistute time-stamped edges. There contexts where network might be considered a reflection underlying dynamics some latent space, whereby nodes associated with locations their relative distances drive interaction tendencies. As time passes can change assuming new configurations, different patterns. The aim this paper is to define space event model. We then develop computationally efficient method for inferring nodes. make use Expectation Maximization algorithm embeds an extension universal Kalman filter. filters known being effective tools context tracking objects successful applications fields geolocalization. extend its application networks by filtering signal from sequence adjacency matrices recovering hidden movements. Besides our formulation includes also more traditional fixed random effects, achieving general model that suit large variety applications.