Stochastic network modeling as generative social science
Christian Steglich, Tom A. B. Snijders
Abstract
Stochastic models of sociocentric networks were originally developed for testing hypotheses about micro-level dependencies on the basis of empirical network data. Such dependencies can lead to phenomena like clustering, hub formation, and network autocorrelation. Due to the complex nature of sociocentric networks, parameter estimates of these models are typically obtained by simulation-based inference. This opens the possibility of re-purposing these models as simulation tools for the study of emergent macro-level phenomena. The combination of _tting micro-level models to empirical data sets and explanation of macro-level outcomes renders these models powerful tools for sociological inquiry into interdependent social systems. In this chapter, the use of stochastic actor-oriented models as generative models for such networked social systems is discussed. This is illustrated with an investigation of the emergence of subgroups in adolescents' friendship networks.