Overview and details of the sessions of this conference. Please select a date or location to show only sessions at that day or location. Please select a single session for detailed view (with abstracts and downloads if available).
In large networks with thousands or millions of actors, the interactions among actors are not affected by the interactions among all other actors, because many social networks are more local than global in nature: Indeed, actors may not even know most other actors, and therefore cannot be influenced by them. A simple class of models that respects the local nature of many social networks assumes that actors are divided into communities and that actors are affected by other actors of the same community, but are not affected by actors outside of the community. The communities may be known or unknown. If the communities are unknown, one can infer the unobserved communities from the observed social network along with the social forces that govern interactions among actors within and between communities. The proposed workshop focuses on next-generation ERGMs for large networks implemented in R package bigergm, which is an evolution of R packages hergm and lighthergm. The workshop will introduce the basic ideas of next-generation ERGMs and will demonstrate them by examples. Participants will be provided with sample R scripts.
Software:
Fritz, Schweinberger, Komatsu, Dahbura, Nishida, and Mele (2024). R package bigergm. https://cran.r-project.org/web/packages/bigergm/index.html
Literature:
The basic idea is introduced in Schweinberger and Handcock (2015). Local dependence in random graph models: Characterization, properties and statistical inference. Journal of the Royal Statistical Society, Series B, 77, 647-676.
An application to systemic risk in social networks can be found in Fritz, Georg, Mele and Schweinberger (2024). Vulnerability webs: Systemic risk in software networks.
Computational details are provided in Babkin, Stewart, Long, and Schweinberger (2020). Large-scale estimation of random graph models with local dependence. Computational Statistics & Data Analysis, 152, 1-19.
Presentations
Next-generation ERGMs: Scaling Up
Michael Schweinberger, Cornelius Fritz
In large networks with thousands or millions of actors, the interactions among actors are not affected by the interactions among all other actors, because many social networks are more local than global in nature: Indeed, actors may not even know most other actors, and therefore cannot be influenced by them. A simple class of models that respects the local nature of many social networks assumes that actors are divided into communities and that actors are affected by other actors of the same community, but are not affected by actors outside of the community. The communities may be known or unknown. If the communities are unknown, one can infer the unobserved communities from the observed social network along with the social forces that govern interactions among actors within and between communities. The proposed workshop focuses on next-generation ERGMs for large networks implemented in R package bigergm, which is an evolution of R packages hergm and lighthergm. The workshop will introduce the basic ideas of next-generation ERGMs and will demonstrate them by examples. Participants will be provided with sample R scripts.
Software:
Fritz, Schweinberger, Komatsu, Dahbura, Nishida, and Mele (2024). R package bigergm. https://cran.r-project.org/web/packages/bigergm/index.html
Literature:
The basic idea is introduced in Schweinberger and Handcock (2015). Local dependence in random graph models: Characterization, properties and statistical inference. Journal of the Royal Statistical Society, Series B, 77, 647-676.
An application to systemic risk in social networks can be found in Fritz, Georg, Mele and Schweinberger (2024). Vulnerability webs: Systemic risk in software networks.
Computational details are provided in Babkin, Stewart, Long, and Schweinberger (2020). Large-scale estimation of random graph models with local dependence. Computational Statistics & Data Analysis, 152, 1-19.
