Testing of and Modeling Dependencies Between a Network and Nodal Attributes

• Activity Number: 668
• Type: Topic Contributed
• Date/Time: Thursday, August 8, 2013 : 10:30 AM to 12:20 PM
• Sponsor: Section on Statistics in Epidemiology
• Author(s): Bailey Fosdick*+ and Peter David Hoff
• Companies: University of Washington and University of Washington
• Keywords: network ; nodal attribute ; dependencies ; covariance ; latent variable

Abstract

Social network analysis is often focused on characterizing the dependencies between network relations and node-level attributes. Potential relationships are typically explored by modeling the network as a function of the nodal attributes or by modeling the attributes as a function of the network. These methods require specification of the exact nature of the association between the network and attributes, reduce the network data to a small number of summary statistics, and are unable provide predictions simultaneously for missing attribute and network information. In this talk we introduce methodology that addresses these shortcomings; we propose a formal testing procedure for determining if a relationship between the network and attributes exists and a joint model for the network and attributes that can capture a variety of dependence patterns. This model is an extension of the latent variable network models presented in Hoff (2005) and Hoff (2009).