***** To join INSNA, visit http://www.insna.org *****
Hi Matthew --
Thanks for this reference. Lots of good material is reviewed in here.
One comment on a statement you make about ERGMs:
Despite significant advances in recent years (e.g. Goodreau (2007)), ERGMs
still suffer from problems of degeneracy and computational intractability
for large network sizes, which has limited their use in epidemic
There seems to be a bit of confusion about ERGMs and degeneracy, which is
not surprising, since it's a complex issue. In general, these problems
(when they exist) are not so much a problem with ERGMs per se -- they are
a problem with our intuition about how to specify an ERGM model to
represent the process we are trying to estimate from data. A good
reference for this is Mark Handcock's “Assessing Degeneracy in
Statistical Models of Social Networks”, CSSS Working Paper no. 39
In any case, degeneracy is not as much of a problem in HIV network
epidemiology for several reasons:
Most sexual networks can be characterized with simple terms for mixing and
degree distributions -- in fact, they almost have to be since we almost
never have data on data on complete sexual networks, and are typcially
constrained to model the structural features observable in egocentric
networks. This is quite different than the friendship nets from Goodreau
et al 2007, which have triad terms for "friend of a friend" effects. And
in that paper, it is worth noting that using the curved exponential family
terms dramatically reduces the problem of degeneracy in estimation.
Large networks are not more problematic than small. See Morris et al.
2009 (AJPH) which estimates from a 15K sample and simulates a 10K node
network. Degeneracy (and computational intractability) was not an issue.
The most recent relevant work is Goodreau et al. 2010 (AIDS and Behavior),
which uses the ERGM framework to estimate and simulate HIV transmission
on a dynamic sexual network. Degeneracy was not an issue here either.
On Tue, 1 Mar 2011, Matthew Vernon wrote:
> ***** To join INSNA, visit http://www.insna.org *****
> Dear Miguel,
> Miguel Valencia wrote:
>> I'm currently working in some aspects about SIR and SI models.
>> However, I haven't been able to find papers or bibliography about
>> applications of these methods. Does anyone know about references I
>> can read about SIR applications? Thank you, I would really appreciate
> If you don't mind me plugging my own work, a review of networks an
> epidemiology which I've been involved in is available as a pre-print:
> It will be appearing in Interdisciplinary Perspectives on Infectious Diseases
> very shortly, but we're at the "approving draft versions from the publishers"
> Matthew Vernon, Research Fellow
> Ecology and Epidemiology Group,
> University of Warwick
> SOCNET is a service of INSNA, the professional association for social
> network researchers (http://www.insna.org). To unsubscribe, send
> an email message to [log in to unmask] containing the line
> UNSUBSCRIBE SOCNET in the body of the message.
Professor of Sociology and Statistics
Director, UWCFAR Sociobehavioral and Prevention Research Core
University of Washington
Seattle, WA 98195-4322
Office: (206) 685-3402
Dept Office: (206) 543-5882, 543-7237
Fax: (206) 685-7419
[log in to unmask]
SOCNET is a service of INSNA, the professional association for social
network researchers (http://www.insna.org). To unsubscribe, send
an email message to [log in to unmask] containing the line
UNSUBSCRIBE SOCNET in the body of the message.