Discussion on the paper by Handcock, Raftery and Tantrum

Tom A. B. Snijders, Tony Robinson, Anthony C. Atkinson, Marco Riani, Isobel Claire Gormley, Thomas Brendan Murphy, Trevor J. Sweeting, David S. Leslie, Nicholas T. Longford, John T. Kent, Tony Lawrance, Edoardo M. Airoldi, Julian Besag, David M. Blei, Stephen E. Fienberg, Ronald L. Breiger, Carter T. Butts, Patrick Doreian, Vladimir Batagelj, Anuška Ferligoj, David Draper, Marijtje A. J. van Duijn, Katherine Faust, Miruna Petrescu‐Prahova, Jonathan J. Forster, Andrew Gelman, Steven M. Goodreau, Priscilla E. Greenwood, Katharina Gruenberg, Brian Francis, Christian Hennig, Peter D. Hoff, David R. Hunter, Dirk Husmeier, C. A. Glasbey, David Krackhardt, Jouni Kuha, Anders Skrondal, Andrew B. Lawson, Tim Futing Liao, Bruno Mendes, Gesine Reinert, Sylvia Richardson, Alex Lewin, D. M. Titterington, Stanley Wasserman, Adriano Velasque Werhli, Peter Ghazal, Ghazal, PUniversity 2007

Abstract

Network models are widely used to represent relations between interacting units or actors. Network data often exhibit transitivity, meaning that two actors that have ties to a third actor are more likely to be tied than actors that do not, homophily by attributes of the actors or dyads, and clustering. Interest often focuses on finding clusters of actors or ties, and the number of groups in the data is typically unknown. We propose a new model, the latent position cluster model, under which the probability of a tie between two actors depends on the distance between them in an unobserved Euclidean 'social space', and the actors' locations in the latent social space arise from a mixture of distributions, each corresponding to a cluster. We propose two estimation methods: a two-stage maximum likelihood method and a fully Bayesian method that uses Markov chain Monte Carlo sampling. The former is quicker and simpler, but the latter performs better. We also propose a Bayesian way of determining the number of clusters that are present by using approximate conditional Bayes factors. Our model represents transitivity, homophily by attributes and clustering simultaneously and does not require the number of clusters to be known. The model makes it easy to simulate realistic networks with clustering, which are potentially useful as inputs to models of more complex systems of which the network is part, such as epidemic models of infectious disease. We apply the model to two networks of social relations. A free software package in the R statistical language, latentnet, is available to analyse data by using the model.
BibTeX
@inproceedings{snijders2007discussion,
  author = {Tom A. B. Snijders and Tony Robinson and Anthony C. Atkinson and Marco Riani and Isobel Claire Gormley and Thomas Brendan Murphy and Trevor J. Sweeting and David S. Leslie and Nicholas T. Longford and John T. Kent and Tony Lawrance and Edoardo M. Airoldi and Julian Besag and David M. Blei and Stephen E. Fienberg and Ronald L. Breiger and Carter T. Butts and Patrick Doreian and Vladimir Batagelj and Anuška Ferligoj and David Draper and Marijtje A. J. van Duijn and Katherine Faust and Miruna Petrescu‐Prahova and Jonathan J. Forster and Andrew Gelman and Steven M. Goodreau and Priscilla E. Greenwood and Katharina Gruenberg and Brian Francis and Christian Hennig and Peter D. Hoff and David R. Hunter and Dirk Husmeier and C. A. Glasbey and David Krackhardt and Jouni Kuha and Anders Skrondal and Andrew B. Lawson and Tim Futing Liao and Bruno Mendes and Gesine Reinert and Sylvia Richardson and Alex Lewin and D. M. Titterington and Stanley Wasserman and Adriano Velasque Werhli and Peter Ghazal and Ghazal, P},
  title = {{Discussion on the paper by Handcock, Raftery and Tantrum}},
  booktitle = {University},
  year = {2007},
}

Loading related content...

Figure