BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Latent Space Stochastic Block Model for Social Networks - Brendan 
 Murphy (University College Dublin)
DTSTART:20160726T103000Z
DTEND:20160726T110000Z
UID:TALK66849@talks.cam.ac.uk
CONTACT:INI IT
DESCRIPTION:A large number of statistical models have been proposed for so
 cial network analysis in recent years. In this paper\, we propose a new mo
 del\, the&nbsp\;latent position stochastic block model\, which extends and
  generalises both latent space model (Hoff et al.\, 2002) and stochastic b
 lock model (Nowicki and Snijders\, 2001). The probability of an edge betwe
 en two actors in a network depends on their respective class labels as wel
 l as latent positions in an unobserved latent space. The proposed model is
  capable of representing transitivity\, clustering\, as well as disassorta
 tive mixing. A Bayesian method with Markov chain Monte Carlo sampling is p
 roposed for estimation of model parameters. Model selection is performed b
 y directly estimating marginal likelihood for each model and models of dif
 ferent number of classes or dimensions of latent space can be compared. We
  apply the network model to one simulated network and two real social netw
 orks.
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR