*****  To join INSNA, visit  *****

Dear Martina,

I also could hardly agree with you more! The IMA workshop that you organized
in Minnesota in 2003 was a huge inspiration to me at a formative period in
my graduate study when I was also taking Steve Borgatti's wonderful class.

One of the papers that I mentioned before in reply to this thread combines
econometric estimation with multi-agent based social simulation:

Dugundji ER, Gulyás L (2008) Socio-dynamic discrete choice on networks:
Impacts of agent heterogeneity on emergent equilibrium outcomes. Environment
& Planning B: Planning and Design 35(6): 1028-1054

This paper considers choice between multiple options.

Another earlier paper also takes this same approach of combining econometric
estimation with multi-agent based social simulation for a binary case:

Dugundji ER, Gulyás L (2005) Sociodynamic discrete choice in an empirical
example on intercity travel demand: an agent-based approach, analytical
benchmark and some issues in estimation. In: Proceedings of the Workshop on
Modelling Urban Social Dynamics, University of Surrey, Guildford

We presented the concept version of the model and initial results two years
earlier at CUPUM to get feedback from the urban planning community:

Dugundji ER, Gulyás L (2003) Empirical estimation and multi-agent based
simulation of a discrete choice model with network interaction effects. In:
Proceedings of the 8th International Conference on Computers in Urban
Planning and Urban Management, Sendai, CD-ROM

In this early work, we found exactly this point that you mentioned about
threshold properties of a complex system. Our results suggest that when a
network has the small-world property for our particular study, the system
behaves in the long run as a much simplified model with global mean field
information, but that sparse networks are more dependent on the actual
reference structure.

Also in the class that I have coordinated, Advanced Network Analysis, at the
University of Amsterdam, we have had a number of guest speakers over the
years addressing in turn experiments, statistics and simulation among other
topics. Hopefully sometime soon if I manage to find some free time I will be
able to resolve technical issues to make these videos available to the
community under a creative commons license on a case-by-case basis as agreed
with the guest speakers.

Best regards,

-----Oorspronkelijk bericht-----
Van: Social Networks Discussion Forum [mailto:[log in to unmask]] Namens
Martina Morris
Verzonden: woensdag 2 maart 2011 23:13
Aan: [log in to unmask]
Onderwerp: Re: Measuring contagion in longitudinal behavior data

*****  To join INSNA, visit  *****

Thanks Sinan,

Two examples from our team are:

Goodreau, S. M., S. Cassels, et al. (2010). "Concurrent partnerships, Acute
Infection and Epidemic Dynamics in Zimbabwe." AIDS and Behavior Online
First, Dec 29, 2010, DOI: 10.1007/s10461-010-9858-x.

Morris, M., A. Kurth, et al. (2009). "Concurrent partnerships and HIV
prevalence disparities by race: Linking science and public health practice."
Amer J Pub Health 99(6): 1023 - 1031.


On Wed, 2 Mar 2011, Sinan Aral wrote:

> *****  To join INSNA, visit  *****
> Martina,
> I couldn't agree more. We make some of these same points and try to 
> take this combined empirical estimation / simulation approach in this 
> paper (although we don't use ERGM):
> Aral, Muchnik and Sundararajan. (2011) "Engineering Social Contagions: 
> Optimal Network Seeding and Incentive Strategies"
> We argue here that simulations based on real data are more realistic 
> and useful and so pursue this strategy using the 30M node Yahoo IM 
> network from our PNAS paper (that I pointed to before) along with data 
> on the day by day adoption of a new mobile service product by the same 
> users - which we use to estimate the probability of diffusion in the 
> system as you suggest below. We then simulate various firm marketing 
> interventions including targeting/seeding and incentives designed to
promote diffusion.
> We should have a slightly updated draft of this paper available very soon.
> I would love it if people could point us to other papers like this too...!
> Best
> SA
> Sinan Aral
> Assistant Professor, NYU Stern School of Business.
> Research Affiliate, MIT Sloan School of Management.
> Personal Webpage:
> SSRN Page:
> WIN Workshop:
> Twitter:
> On 3/2/2011 1:52 PM, Martina Morris wrote:
>> Entering this conversation late, so I may have missed some things, 
>> but I wanted to put a plug in for using a combined empirical + 
>> simulation based approach to the problem, in case this hasn't been
mentioned before.
>> If diffusion is thought of as a combination of network 
>> connectivity/structure, transmission probability, and duration of 
>> "infectiousness" (this is the classic formulation in epidemiology), 
>> there are quite a few parameters that can determine the dynamics of 
>> the diffusion.  Unless you're interested in a black-box type of 
>> projection, understanding how each of these components influences the 
>> process is an important goal.
>> Simulation of the process (before getting into the empirical 
>> trenches) allows for quite a bit of exploratory analysis and theoretical
>> The factors that influence the formation/dissolution of ties in the 
>> network may or may not be different than the factors that influence 
>> transmission across the ties.  But the impact of these two types of 
>> components on diffusion dynamics is quite different.  
>> Presence/absence of ties influence the connectivity of the network, 
>> which has the usual non-linearities and threshold properties of a 
>> complex system (think emergence of the giant component), and this can 
>> become a key limiting factor in sparse networks (sexual networks are the
paradigmatic example).
>> For those of us interested in the role of network structure, it is 
>> important that the model used to simulate the network is one that can 
>> also be used to estimate the model parameters from data.  That's one 
>> of the things that makes ERGMs attractive -- they do this double duty 
>> for estimation and simulation, allowing you to simulate networks that 
>> match the observable summary network statistics from your data.
>> You would also need empirical estimates of the probability of 
>> diffusion and the "duration of infectiousness"  to simulate the whole
>> In contrast to a single realization of the stochastic diffusion 
>> process -- which is what most empirical studies would give you (RCT 
>> or observational)
>> -- simulating the network (with empirically estimated parameters for 
>> network dynamics and diffusion parameters) allows you to map out the 
>> distribution of outcomes from the system, and to see how sensitive it 
>> is to small changes in the underlying parameters.  Arguably, a lot 
>> more insight into the phenomenon of interest.
>> best,
>> Martina

  Professor of Sociology and Statistics
  Director, UWCFAR Sociobehavioral and Prevention Research Core
  Box 354322
  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 ( To unsubscribe, send an email
message to [log in to unmask] containing the line UNSUBSCRIBE SOCNET in
the body of the message.

SOCNET is a service of INSNA, the professional association for social
network researchers ( To unsubscribe, send
an email message to [log in to unmask] containing the line
UNSUBSCRIBE SOCNET in the body of the message.