Print

Print


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

Dear Thomas,

I support what Xi writes. Read the Siena manual for more details!

Further, you should note that 650 nodes with 8000 or 10,000 ties is a
lot for Siena. It will take a lot of computing time; and to make the
analysis meaningful you should have a lot of information concerning
shared contexts, attributes, etc., as predictors for ties. We are
working on a model that is more plausible also for larger networks but
it will be a while before this is available.

Apparently you have been using the older version Siena 3. Currently
Siena is available as an R package, which also allows use of multiple
processors, which will of course speed up computations. Information
about the R version of Siena is at the Siena website,
http://www.stats.ox.ac.uk/siena/

Best wishes,

Tom


Xi Zhu wrote:
> ---------------------- Information from the mail header -----------------------
> Sender:       Social Networks Discussion Forum <[log in to unmask]>
> Poster:       Xi Zhu <[log in to unmask]>
> Subject:      Re: Problems with StOCNET, SIENA Module
> -------------------------------------------------------------------------------
> 
> --0016e6d7e744c6a93b047b6eb124
> Content-Type: text/plain; charset=ISO-8859-1
> 
> *****  To join INSNA, visit http://www.insna.org  *****
> 
> Hi Thomas,
> 
> As I know, StOCNet only takes adjacency matrices as input data. This means
> that you need to recode the edgelist into adjacency matrices saved in ASCII
> format (dat or txt files). Each network should be presented by n lines with
> n integer numbers separated by blanks, and each line is ended by a hard
> return. No node name should appear in an adjacency matrix. If you code node
> names as integers, they will be considered as representing relationships.
> 
> Since you want to do longditudinal network analysis (using SIENA), you also
> need to save each observation of the network (e.g., network of each year) as
> a separate file. StOCNet will allow you to use multiple observations in one
> analytical session.
> 
> You may want to read StOCNet and SIENA manuals. You can find them on StOCNet
> website's Downloads section. Here is the URL link:
> http://stat.gamma.rug.nl/stocnet/
> 
> Hope this would help.
> 
> Best,
> Xi
> 
> On Wed, Dec 23, 2009 at 3:58 AM, Thomas Plotkowiak <[log in to unmask]> wrote:
> 
>> ***** To join INSNA, visit http://www.insna.org ***** Hi,
>>
>> A few members of the SOCNET mailinglist pointed me to SIENA for
>> longitudonal social network analysis. I don't know if my posting is right
>> here, but I am having problems importing two nets of the size of approx 650
>> nodes into the module.
>>
>> When I do the SIENA data specification and add those two networks into
>> "Dep. networks in seq. order" either nothing happens if I click ok, or I am
>> getting the error "Access violation at adress 004E2E40 in module
>> "StOCNet.exe. Read of adress 00DF0008."
>>
>> My data is saved in the edgelist format which is available since v. 4 and
>> looks like this:
>>
>> BoehningB,michaeladam1984,1
>> Elke_Ferner,michaeladam1984,1
>> kaigersch,michaeladam1984,1
>> guentergloser,michaeladam1984,1
>> hubertus_heil,michaeladam1984,1
>> heinzpaula,michaeladam1984,1
>> ...
>> approx 8k rows.
>>
>> Should i re-encode the node names into integers, or ist the edgelist
>> support only beta? Maybe somebody had some luck with this.
>>
>> Cheers
>> Thomas
>>
>>
>> ------
>> Thomas Plotkowiak
>> Research Assistant
>> Institute for Media and Communications Management
>> St. Gallen
>> _____________________________________________________________________
>> 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.
> 
> 
> 
> 

-- 
Tom A.B. Snijders
Professor of Statistics in the Social Sciences
University of Oxford

Professor of Statistics and Methodology
Department of Sociology, University of Groningen

for addresses and telephone numbers: see
http://stat.gamma.rug.nl/snijders/

_____________________________________________________________________
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.