Print

Print


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

Dear Shahadat,

Some experts will probably give you more details, but here is my 
experience with SAOM:

To give you an idea of the technical limitations, I have successfully 
run a model with 1200 actors in less than 24 hours using RSiena (My 
machine: Processor 3.06 Ghz; 16 GB of RAM; 64-bit OS). Then the 
computing time increases exponentially.

Besides the technical limitations, and depending on the kind of network 
you analyze, you might also violate the assumption that each actor knows 
the particular way ties are organized in the network (Snijders et al. 2010).

I also think that a new type of SAOM for the modeling of large networks 
is under development. This should partially overcome the technical 
limitations related to the size of the network, and also relax the 
assumption that actors are able to evaluate their position in the 
network (I think that the idea is to switch from the assumption of 
information on the entire structure to information on the local 
structure, which is also more realistic for small networks anyway).

Best,

Pierre-Alex

Reference:

Snijders T., Van De Bunt G. and Steglich C. (2010) Introduction to 
actor-based models for network dynamics, Social Networks 32, 44 -- 60.


On 14-2-2012 11:35, Shahadat Uddin wrote:
> ***** To join INSNA, visit http://www.insna.org *****
>
> Dear All,
>
> Can anyone help me with the answer of the following question:
>
> What is the largest size of a network (in respect of adjacent matrix) 
> so that the known methods (i.e., ERGM  and stochastic actor-level 
> method) can be applied using any available software (e.g., Pnet, R) to 
> capture the network-level dynamics?
>
> Regards
>
> Shahadat
>
> _____________________________________________________________________ 
> 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. 


-- 
Dr. Pierre-Alexandre Balland

Research fellow
Urban and Regional Research Centre Utrecht (URU)
Section of Economic Geography
Faculty of GeoSciences, Utrecht University

Heidelberglaan 2 - Room 625
PO Box 80 115
3508 TC Utrecht, The Netherlands
Phone: +31 (0) 30 253 2726

Homepage: http://econ.geo.uu.nl/balland/balland.html


_____________________________________________________________________
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.