My Photo

« Comparing TV and Blogosphere Mentions | Main | Battelle's Prediction for Search »

January 03, 2010

Comments

Ericness

Hey Matthew,

I have watched this thing several times since you posted it and at first my thought was it was pretty cool even if it is just a power law. But, after seeing it a couple of times it makes me really wonder about some of the conclusions too - namely how much meaning can you really derive from the slope. Just an example - Gourley says that with a higher alpha comes "Fragmented, fluid, more groups, weaker" but if you look the conflict with the highest alpha you have Northern Ireland and if you look at the time series of alpha towards the end of the presentation you MAY have some fairly weird results for the surge.

Also came across these two items you might like that where posted by someone else:
Variation of the Frequency of Fatal Quarrels With Magnitude. Lewis F. Richardson.
http://www.uvm.edu/~pdodds/research/papers/others/1949/richardson1949a.pdf

On the Frequency of Severe Terrorist Events
http://arxiv.org/abs/physics/0606007

Seoroi

He says you need to look at group size to solve the conflict. But his data disproves that in that it bounced around. Seems his analysis of the stats isn't very valuable imho.

The comments to this entry are closed.

Twitter Updates

    follow me on Twitter

    May 2015

    Sun Mon Tue Wed Thu Fri Sat
              1 2
    3 4 5 6 7 8 9
    10 11 12 13 14 15 16
    17 18 19 20 21 22 23
    24 25 26 27 28 29 30
    31            

    Categories

    Blog powered by Typepad