Improving Simulation-Based Algorithms for Fitting ERGMs
Authors:
- 3 April 2012
- journal article
- research article
- Published by Taylor & Francis in Journal of Computational and Graphical Statistics
- Vol. 21 (4) , 920-939
Abstract
Markov chain Monte Carlo methods can be used to approximate the intractable normalizing constants that arise in likelihood calculations for many exponential-family random graph models for networks. However, in practice, the resulting approximations degrade as parameter values move away from the value used to define the Markov chain, even in cases where the chain produces perfectly efficient samples. We introduce a new approximation method along with a novel method of moving toward a maximum likelihood estimator (MLE) from an arbitrary starting parameter value in a series of steps based on alternating between the canonical exponential-family parameterization and the mean-value parameterization. This technique enables us to find an approximate MLE in many cases where this was previously not possible. We illustrate these methods on a model for a transcriptional regulation network for E. coli, an example where previous attempts to approximate an MLE had failed, and a model for a well-known social network dataset involving friendships among workers in a tailor shop. These methods are implemented in the publicly available ergm package for R, and computer code to duplicate the results of this article is included in the online supplementary materials.Keywords
This publication has 21 references indexed in Scilit:
- On the geometry of discrete exponential families with application to exponential random graph modelsElectronic Journal of Statistics, 2009
- A framework for the comparison of maximum pseudo-likelihood and maximum likelihood estimation of exponential family random graph modelsSocial Networks, 2009
- Likelihood inference in exponential families and directions of recessionElectronic Journal of Statistics, 2009
- Goodness of Fit of Social Network ModelsJournal of the American Statistical Association, 2008
- Exploring biological network structure using exponential random graph modelsBioinformatics, 2007
- Curved exponential family models for social networksSocial Networks, 2007
- Inference in Curved Exponential Family Models for NetworksJournal of Computational and Graphical Statistics, 2006
- Network motifs in the transcriptional regulation network of Escherichia coliNature Genetics, 2002
- Markov GraphsJournal of the American Statistical Association, 1986
- Random GraphsThe Annals of Mathematical Statistics, 1959