SPEAKER: Mark Handcock, UCLA
TITLE: “Some new models for social networks”
ABSTRACT: Random graphs, where the connections between nodes are considered random variables, have wide applicability in the social sciences. Exponential-family Random Graph Models (ERGM) have shown themselves to be a useful class of models for representing complex social phenomena.
In this talk we will consider some new classes of models that generalize ERGM in different ways. First, we model the attributes of the social actors as random variates, thus creating a random model of both the relational and individual data, which we call Exponential-family Random Network Models (ERNM). This provides a framework for expanded analysis of network processes, including a new formulation for network regression where the outcomes, covariates and relations are socially endogenous. We illustrate this with a new class of latent cluster models and network regression.
Next we introduce a class of models we call Tapered Exponential-family Random Network Models (TERNM). These models remove the degeneracy properties that hamper ERGM and ERNM while retaining there advantages. We show how these models can provide good fits to large networks.
Finally we introduce spatial temporal exponential-family of point processes (STEPP) models to jointly represent the co-evolution of social relations and individual behavior in discrete time.
This is joint work with Ian E. Fellows and Joshua D. Embree .
Speaker's Website: http://www.stat.ucla.edu/~handcock/
DATE: Thursday, May 16th, 4:10pm
LOCATION: MSB 1147, Colloquium Room
REFRESHMENTS: 3:30pm MSB 4110 (4th floor lounge)
STA 290 Seminar List: https://statistics.ucdavis.edu/seminars