***** To join INSNA, visit http://www.insna.org *****
Hi,
I am not sure that I understand your problem, but I would simply measure the
size of the largest component in each of the networks. If it is above 50% in
all networks, then your hypothesis is confirmed. You may want to define a
stricter threshold, though.
Best regards,
-- Laszlo
--
Laszlo Gulyas, Phd
director
Intelligent Applications and Web Services
AITIA International, Inc.
----- Original Message -----
From: "Andrej Kastrin" <[log in to unmask]>
To: <[log in to unmask]>
Sent: Wednesday, June 15, 2011 7:01 PM
Subject: [SOCNET] Test hypothesis about giant component
> ***** To join INSNA, visit http://www.insna.org *****
>
> Dear all,
>
> I have a sample of 200 independent networks and I want to test the
> hypothesis that majority of vertices belong to one giant component. I
> wonder what is the appropriate approach to do that.
>
> I crawled the Google Scholar, but I didn’t find any pointers about my
> problem. Could I use standard chi-square goodness-of-fit test, where the
> first set of frequencies counts size of giant component (i.e., number of
> vertices in a giant component) and the second set counts the size of the
> rest components (i.e., total number of vertices in a sample – number of
> vertices in a giant component)?
>
> Any suggestions would be greatly appreciated.
>
> Best, Andrej
>
> _____________________________________________________________________
> 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.
|