A0483
Title: Multi-view dynamic social network modeling
Authors: Shun Hin Chan - The Hong Kong University of Science and Technology (Hong Kong) [presenting]
Amanda Chu - The Education University of Hong Kong (China)
Mike So - The Hong Kong University of Science and Technology (Hong Kong)
Abstract: A flexible multi-view dynamic social network model is developed using a regression-like structure, incorporating exogenous and endogenous variables from the lagged networks to model edge changes. The model does not rely on latent space, simplifying network estimation and prediction. Furthermore, it integrates a multi-view feature to represent various relationship types at each time point. The proposed model offers an intuitive interpretation of the estimation. Bayesian model averaging method is also applied to predict networks.
