Light refreshments will be provided.
Tutorial Summary
Exponential Random Graph Models (ERGMs) are a family of statistical models for analyzing networks. ERGMs are an exciting class of models that do not assume the independence of nodes and edges, and allow models to include node attributes, edge attributes, as well as more complex structural features. However, the meaningful application of ERGMs to network data requires problems be conceptualized in a relational way. Moreover, results can be tricky to interpret and sometimes converge to degenerate solutions.
In the first hour of the workshop, we will review the fundamental components and diagnostics of ERGMs. In the second half, we will collectively fit and diagnose ERMGs for two network data sets. The first data set is a network of social distances between occupations in Japan (Hiroki Takikawa - Sociology) and the second is of food exchange in an arctic community (Elspeth Ready - Anthropology).
Resource Links