Information Greedy 'Patchwork' Bootstrap on Random Networks

• Activity Number: 154
• Type: Invited
• Date/Time: Monday, August 10, 2015 : 10:30 AM to 12:20 PM
• Sponsor: SSC
• Author(s): Yulia R. Gel* and Vyacheslav Lyubchich and Lilia L. Ramirez Ramirez
• Companies: The University of Texas at Dallas and University of Maryland Center for Environmental Science and Instituto Tecnológico Autónomo de México
• Keywords: bootstrap ; resampling ; nonparametrics ; random networks ; random graphs ; sampling

Abstract

In this talk we propose a new nonparametric ``patchwork'' resampling approach to network inference based on the adaptation of ``blocking'' argument, developed for bootstrapping of time series and re-tiling for spatial data, to random networks. We discuss how the new ``patchwork'' procedure can be used to quantify estimation uncertainty for network statistics that are functions of degree distribution. We develop a new computationally efficient and data-driven cross-validation algorithm for selecting an optimal ``patch'' size. We illustrate utility of the new ``patchwork'' bootstrap procedure for inference on simulated networks, flight and Wikipedia networks.