Introduction to Statistical Models for Networks

Abstract

The purpose of this short exposition is to discuss the developments in statistical models for networks that have occurred over the past five years,? since the publication of the statistical chapters (8, 9, 10, and 11) of Carrington, Scott, and Wasserman (which were written in 2002). The statistical modeling of social networks is advancing quite quickly. The many exciting new developments include, for instance, longitudinal models for the co-evolution of networks and behavior (Snijders, Steglich & Schweinberger, 2010) and latent space models for social networks (Hoff, Raftery & Handcock, 2002). We do not intend to review all the recent advances but rather limit our scope to a few developments that we have worked on.