***** To join INSNA, visit http://www.insna.org ***** Kevin,

Here are a few links to a related body of work:

http://www.cmu.edu/joss/content/articles/volume12/McCullohCarley.pdf

http://www.dtic.mil/cgi-bin/GetTRDoc?AD=ADA506151

http://www.dtic.mil/cgi-bin/GetTRDoc?AD=ADA497468

http://www.aisb.org.uk/convention/aisb09/Proceedings/SNAMAS09/FILES/McCullohI.pdf

There are two general approaches for running simulations of change over time.  Multi-agent simulation is often used, where you specify a model with agent defined attributes and interaction rules.  I recommend Kathleen Carley's work at www.casos.cs.cmu.edu.

An alternate approach is to use an actor-oriented model.  The advantage here (as described in one of the papers above) is that you can measure significant factors driving network behavior from empirical data and then implement change in an actor-oriented simulation.  I did an Army Technical Report with Josh Lospinoso on this in 2010, however, I could not find it...sorry.

Hope this is of some help. 

Ian McCulloh, Ph.D.

On Sat, Nov 24, 2012 at 1:32 AM, Kevin Hallgren <[log in to unmask]> wrote:
***** To join INSNA, visit http://www.insna.org ***** Hi everyone,

I am interested in running simulations of social network changes over time.  I would like the simulations to include changes in social ties and changes in actor covariates/attributes over time, and ideally the changes in ties and attributes of actors would be able to be influenced by the current social ties and attributes of the actors (i.e., making ties with actors who have similar attributes, and changing attributes to become similar to other actors with social ties). 

Does anyone know of any references or starting places that describe mathematical or programmatic guides for doing this? 

I was thinking of using a stocahstic actor-based modeling approach with unweighted, directional ties (a la Snijders, van de Bundt, & Steglich, 2010), but could use other types of graphs (e.g., with non-directional ties) as long as it allows the mutual influence of changes in social ties and changes in actor covariates to be modeled over time.  I would be interested incorporating some common network effects like transitivity, reciprocity (if using directional ties), outdegree, and a few others that will create networks that could have some plausible resemblance to real-world networks. 

Any references, program manuals, names to look into, etc. are greatly appreciated.

Thanks!
Kevin
[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.

_____________________________________________________________________ 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.