Data-Driven Models for Dynamic Networks in Changing Populations

• Activity Number: 529
• Type: Contributed
• Date/Time: Wednesday, August 4, 2010 : 10:30 AM to 12:20 PM
• Sponsor: Social Statistics Section
• Author(s): Pavel N. Krivitsky*+ and Mark Stephen Handcock and Martina Morris
• Companies: Carnegie Mellon University and University of California, Los Angeles and University of Washington
• Keywords: ERGM ; dynamic networks ; egocentric data ; simulation

Abstract

Models of dynamic networks --- networks that evolve over time --- have manifold applications. Applications in epidemiology present their own challenge, in that the data available are often limited to egocentric views of the network processes, and in that evolution of these networks comprises not only changes in relationship states but also changes in attributes and numbers of actors. We develop and apply a discrete-time generative model for social network evolution that seeks to inherit the richness and flexibility of exponential-family random graph models, while adjusting for changing network size and composition, and facilitate modeling of tie duration distributions. We develop methods to fit the model to available cross-sectional, egocentric network and/or tie duration data, and present an application to sexual partnership data in the context of modeling the structure of HIV spread.