Dear Edmund et al.,
The segregation model in NetLogo's Models Library can easily be modified to
provide this functionality. (See the attached .nlogo file.)
I haven't much explored what the effect is of having more colours. (Use
BehaviorSpace to conduct the experiments you want.) However, a quick play
with it reveals that when agents want 30% of neighbours to be similar, the
final percent similar is 80% with 12 colours, compared with 72% with 2
colours. Also, the time (ticks) required for the system to reach a steady
state grows longer with the number of colours.
Hope this matches the results from other models and papers!
Christopher Watts
____________________________________________________
Dr Christopher J Watts
Senior Scientist
Chair in Human-Environment Relations, Department of Geography
Ludwig-Maximilians University, Munich
Germany
[log in to unmask]
http://www.geographie.uni-muenchen.de/department/fiona/personen/index.php?pe
rsonen_details=1&user_id=211
Post address:
Lehrstuhl für Mensch-Umwelt-Beziehungen
Luisenstraße 37
80333 München
New book:
Simulating Innovation: Computer-based Tools for Rethinking Innovation
http://www.e-elgar.co.uk/bookentry_main.lasso?id=13981
____________________________________________________
-----Ursprüngliche Nachricht-----
Von: News and discussion about computer simulation in the social sciences
[mailto:[log in to unmask]] Im Auftrag von Edmund Chattoe-Brown
Gesendet: Dienstag, 2. September 2014 16:39
An: [log in to unmask]
Betreff: [SIMSOC] Three (Plus) Type Schelling
Dear All,
Does anyone know a published Schelling segregation model variant with 3 or
more types? I don't recall ever seeing one and I am coming up empty on
google. (The search terms are too weak ...)
Will summarise back to the list ...
Thanks,
Edmund
--
Edmund Chattoe-Brown
[log in to unmask]
--
http://www.fastmail.fm - IMAP accessible web-mail
|