Exponential random graph models for social networks, using Pnet

Garry Robins, Peng Wang, Johan Koskinen & Dean Lusher
University of Melbourne, Australia, University of Manchester & Swinburne University of Technology
garrylr@unimelb.edu.au

LIMITED to 30 Seats

This workshop is in conjunction with the release of a new text on ERGMs: Lusher, D., Koskinen, J., & Robins, G. (2012). Exponential random graph models for social networks: Theory, methods and applications. Cambridge University Press.

The workshop will introduce and utilize PNet, simulation and estimation software for ERGMs, including the new software for multilevel networks, MPNet.

The general theoretical background to ERGMs will be reviewed, model formulation discussed, and simulation, estimation and goodness of fit procedures presented. Meaningful interpretation of parameter estimates will be emphasized, as well as the importance of model goodness of fit. Social selection models incorporating individual-level attributes will also be described. For those who bring laptops to the workshop, hands-on exercises using PNet will be included, both in simulating networks and in fitting illustrative data sets. Extensions to models for multiple networks, bipartite networks and social influence processes on networks may also be presented. New models for multilevel networks will be introduced, using MPnet.

Participants will be expected to have prior knowledge of fundamental social network concepts and terminology, and to have some knowledge of standard statistical ideas (e.g. the notion of a statistical distribution, regression, logistic regression).

After completing the workshop we expect participants to:

  • have a good understanding of the principles and aims of ERGM methodology;
  • be aware of common obstacles and misconceptions with ERGMs;
  • have a working handle on the workflow, from theoretical specification, technical specification and model fitting and interpretation and improvement;
  • be familiar with and have a readied capacity for picking up extensions and more advanced topics;
  • feel confident in being able to fit ERGMs to their own data

*PNet is freely available to participants (www.sna.unimelb.edu.au). Please note it is written for Windows PCs and may not function well under Mac or Linux operating systems. (However, PNet usually works well in a non-native windows environment, such as parallels or boot camp.)
Other Background Reading: Robins, G., Pattison, P., Kalish, Y., & Lusher, D. (2007). An introduction to exponential random graph (p*) models for social networks. Social Networks, 29, 173-191.
Robins, G.L., Snijders, T.A.B., Wang, P., Handcock, M., & Pattison, P. (2007). Recent developments in exponential random graph (p*) models for social networks. Social Networks, 29, 192

Advertisements