[p2p-hackers] zipf's law

[Michael Rogers]

Hi John,

You might also be interested in this paper:

http://www.eecs.harvard.edu/~michaelm/NEWWORK/postscripts/history.pdf

The crux of it is that lognormal and Zipf distributions can be generated 
using very similar multiplicative growth models.

Cheers,
Michael

John Casey wrote:

>Dear All,
>
>I was wondering if any one might help explain what is the best way to
>generate random variables that are distributed according to a zipf
>distribution with a slope of -1 and an output range of 1 to 30. I have
>tried using cern's cern.jet.random.Distributions nextZipfInt(... )
>(see http://hoschek.home.cern.ch/hoschek/colt/V1.0.3/doc/cern/jet/random/Distributions.html)
>but the output of this generator is severely clipped if I attempt to
>change the output range. I have also tried generating a series of data
>points that are derived using zipf's law and have been able to
>uniformly at random sample these values. This method seems to work
>quite, and the distribution of sampled values better approximates
>zipf's law than using the cern random generator. In terms of
>methodology is this method acceptable?? it seems to be a bit of a
>kludge to me. But I am having difficulty finding a better solution.
>Any ideas ?? thanks.
>_______________________________________________
>p2p-hackers mailing list
>p2p-hackers at zgp.org
>http://zgp.org/mailman/listinfo/p2p-hackers
>_______________________________________________
>Here is a web page listing P2P Conferences:
>http://www.neurogrid.net/twiki/bin/view/Main/PeerToPeerConferences
>  
>