University of Washington - Statistics
Latent variable network models provide low-dimensional representations of relational patterns in terms of additive and multiplicative actor-specific effects. In this talk we discuss these models in two contexts. First, we extend this class of models to estimate and make inference on the dependencies between a set of network relations and actor-specific attributes. Approaches to this problem typically condition on either the relations or attributes and are unable to provide predictions simultaneously for missing attribute and network information. We propose methodology for a united approach to analysis that allows for testing for dependencies between the relations and attributes, and in the event the test concludes such structure exists, jointly modeling the relations and attributes to conduct inference and make predictions for missing values. Second, we discuss Bayesian estimation procedures for a general class of these latent variable network models. We propose modifications to a basic Markov chain Monte Carlo algorithm that significantly improve the efficiency of the sampler and illustrate the inadequacies of a mean-field variational approach for this model class.