Print

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

I recommend using ORA.  ORA is pretty easy to figure out and it can import your UCINet files.  ORA is available at www.casos.cs.cmu.edu.  Once you get the program and start it, you import files by clicking the magic wand at the upper right.  The easiest options are to run the QAP/MRQAP Report, which will give you correlation, Hamming distance (if dichotomous) and Euclidean distance.  The other option is to click on the tool bar at the top and find the option for "View Network Over Time."  This feature has many tools for over time analysis.  You have the simple ability to view measures over time, you can apply Fourier analysis to detect periodicity in your data, you can also apply statistical process control to detect changes over time.

Another option is to load your networks in R (not sure how the import works for that one).  You can run an ergm on one network and use the edgecov(network2) which will enable you to test in an ergm another network covariate.  You can actually include as many edgecov terms as you like corresponding to other networks, but there may become a degeneracy issue.  If the edgecov term is significant, that would indicate a significant relationship between the networks given any other terms you include in the model.  This approach is a bit more complex and I am not sure it gets at what you are after.

RSiena also provides some overtime analysis options, depending on what you are trying to do.

Ian

Ian McCulloh
Major, US Army
Assistant Professor
U.S. Military Academy
West Point, NY 10996

On Wed, Dec 23, 2009 at 6:48 AM, Kathleen Carley <[log in to unmask]> wrote:
*****  To join INSNA, visit http://www.insna.org  *****

There are many procedures in ORA for this - both for same node,
overlapping node, and distinct node networks

Lietz, Haiko wrote:
> ***** To join INSNA, visit http://www.insna.org *****
>
> Dear Thomas,
>
>
>
> If your node set is the same for each point in time, you can easily use
> the matrix algebra environment in Ucinet. But if that’s not the case I
> don’t know how to do it in either Ucinet or Pajek. What would be the
> best option then? NetworkX in Python? R?
>
>
>
> Best
>
>
>
> Haiko
>
>
>
>
>
> ------------------------------------------------------------------------
>
> *Von:* Social Networks Discussion Forum [mailto:[log in to unmask]]
> *Im Auftrag von *Thomas Plotkowiak
> *Gesendet:* Montag, 21. Dezember 2009 18:03
> *An:* [log in to unmask]
> *Betreff:* [SOCNET] Diff of two networks
>
>
>
> ***** To join INSNA, visit http://www.insna.org ***** Hi,
>
> I've got a technical question, which i think can be solved with ucinet,
> but i don't know how. Maybe there are other tools for this.
> I would like to make a diff of two networks in ucinet. I have taken a
> snapshot of 700 nodes and their 10000 connections on one day, and then
> did so subsequently for 40 days.
> I would like to diff the network of one day with the next day, to see
> which connection were created between which nodes.
>
> Is there a smart way to do this?
>
> Cheers
> Thomas
>
> _____________________________________________________________________
> 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.

_____________________________________________________________________
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.