Exponential random graph models for little networks

Statistical models for social networks have enabled researchers to study complex phenomena that give rise observed patterns of relationships among actors and gain a rich understanding the interdependent nature ties actors. Much this research has focused on within medium large groups. To date, these advances in statistical networks, particular, Exponential-Family Random Graph Models (ERGMS), rarely been applied small despite network data teams, families, personal being common many fields. In paper, we revisit estimation ERGMs propose using exhaustive enumeration when possible. We developed an R package implements pooled Maximum Likelihood Estimation (MLE), called “ergmito”. Based results extensive simulation assess properties MLE estimator, conclude there are several benefits direct compared approximate methods creates opportunities valuable methodological innovations can be modeling with ERGMs.