Re: [igraph] Help in power.law.fit

[Gábor Csárdi]

Dear Anupam,

On Thu, Sep 10, 2009 at 2:43 PM, anupam sinha <address@hidden> wrote:
> Dear all,
>              Apologies for the long mail I am working on directed metabolic
> networks where enzymes form the nodes of the networks.I am  calculating the
> beweenness centralities of all the nodes. It is known that the Betweenness
> centrality values of a network follow a power law distribution.
>
>>sce_met_bet<-betweenness(sce_met)
>> sce_met_bet
[...]
>
>> power.law.fit(sce_met_bet)
> Error in optim(start, f, method = method, hessian = TRUE, ...) :
>   initial value in 'vmmin' is not finite
>
>
>
>  I have a couple of queries
>
> 1 . Is it appropriate for me to  fit power law distribution to this dataset?

Maybe, but not with the power.law.fit() function, because that is for
count data. You can simply use the formula given in the paper that is
referred on the power.law.fit manual page.

> 2. Why am I getting this error?

Because there is a zero in you vector, and you cannot have that from a
power-law distribution.

Best,
Gabor

> _______________________________________________
> igraph-help mailing list
> address@hidden
> http://lists.nongnu.org/mailman/listinfo/igraph-help
>
>



-- 
Gabor Csardi <address@hidden>     UNIL DGM