Title: Novel network change point methods
Authors: Ivor Cribben - Alberta School of Business (Canada) [presenting]
Abstract: Identifying change points in dynamic network structures has become increasingly popular across various domains, from neuroscience to telecommunication to finance. We will present two new change point detection methods. The first method uses non-negative matrix factorization, an unsupervised dimension reduction technique, and a new binary search algorithm to identify multiple change points. The second method considers changes in vine copula structure, various state-of-the-art segmentation methods to identify multiple change points, and a likelihood ratio test or the stationary bootstrap for inference. The vine copulas allow for various forms of dependence. We apply both methods to simulated, financial and to functional magnetic resonance imaging (fMRI) data sets.