Just one for what it's worth:
how (philosphically and practically) to deal with the issues of
autocorrelation that are a characteristic of the whole field. My hunch is
that all social processes are autocorrelated (in the real world) - ie. that
is a part of what society is about, but statisticians seem to have a
'problem' with this. Unpackingthe history of that problem and its various
manifestations in social simulation (and the social sciences in general)
might provide a basis for a Masters project. And I'd like to see the result!
Alan Penn
UCL
> -----Original Message-----
> From: News and discussion about computer simulation in the social
> sciences [mailto:[log in to unmask]]On Behalf Of Steve Wilcox
> Sent: 03 August 2002 05:17
> To: [log in to unmask]
> Subject: Statistical research issues
>
>
> Pardon the intrusion of a lurker, but the temptation to ask a question is
> just too great. I am considering Master's thesis topics in Statistics and
> would like solicit suggestions of interesting topics in the interface
> between Statistics and {computational and mathematical
> organization theory,
> computational economics, and/or complex adaptive systems}. What are the
> important unsolved statistical issues that arise in social science
> simulations?
>
> Thank you for your consideration of this question.
>
> Steve Wilcox
> [log in to unmask]
>
|