A Stochastic Actor Oriented approach to model policy-driven innovation network dynamics

Stochastic actor oriented models (SAOM) are of growing importance to study network dynamics underlying the theoretical micro mechanisms that induce the evolution of relations among a set of social actors. SAOM represents a suitable methodological framework to investigate the evolution of policy driven innovation networks among heterogeneous agents. The empirical application focuses on a set of policies that support networking among SMEs and other partners implemented by the Tuscany region (Italy) in the programming period 2000/2006. We consider a set of about 350 agents involved in 168 funded projects of variable duration ranging from 4 up to 18 months. Model specification accounts for several structural hypotheses on strategic partner choice in undirected networks (transitivity, popularity, homophily), controlling for agents characteristics (e.g. agent type, localisation) and project duration. In order to realistically frame how the decision to create or dissolve collaboration ties is taken, we consider either one-sided or two-sided initiative models.