B0611
Title: Dependent structures in network data
Authors: Sharmodeep Bhattacharyya - Oregon State University (United States) [presenting]
Shirshendu Chatterjee - City University of New York (United States)
Soumendu Sundar Mukherjee - Indian Statistical Institute (India)
Abstract: Statistical analysis of networks generated from exchangeable network models has been extensively studied in the literature. One primary property of exchangeable network models is the conditional independence of edge formation. We extend the framework of network formation to include dependent edges with an emphasis on generating networks with all five properties of sparsity, small-world, community structure, power-law degree distribution, and transitivity or high triangle count. We propose a class of models, called Transitive Inhomogeneous Erdos-Renyi (TIER) models, which we show have all five properties. We also perform inferential tasks, such as parameter estimation, community detection, and change-point detection for sequences of dependent networks from Inhomogeneous Erdos-Renyi (IER) and TIER models. We validate our results using simulation studies too. If time permits, we will talk about some recent developments in the estimation of the number of communities using Bethe Hessian matrices.
