Re: [igraph] errors using power.law.fit

[Gábor Csárdi]

On Tue, Nov 16, 2010 at 11:28 AM, Tamas Nepusz <address@hidden> wrote:
>> Which is quite surprising, because there is no exact formula for
>> discrete data AFAIK.
> Correction: no, they don't calculate alpha directly, they just simply
> try every possible alpha with some granularity in a given range,
> calculate the likelihood, and then choose the alpha which yields the
> maximal likelihood.

Well, then this is an optimization algorithm after all, a quite rude one. :)

Gabor

-- 
Gabor Csardi <address@hidden>     UNIL DGM