fitting exponential and power law distributions in R and comparing better fit

Question

I have a network which I need to fit with power law distribution and exponential distribution and compare them, choosing the better fit.

I have degree distribution data retrived using igraph package degree.distribution function:

 degree.distribution (data, mode = "all", cumulative = FALSE)

which returns results vector e.g.

0.000000000 0.289772727 0.278409091 0.159090909 0.204545455 0.051136364 0.005681818 0.011363636

I tried fitting it using igraph's power.law.fit as below:

power.law.fit(deg.dist, impelementation = "plfit")

$continuous
[1] TRUE

$alpha
[1] 1.493625

$xmin
[1] 0.008

$logLik
[1] 10.32315

$KS.stat
[1] 0.1248314

$KS.p
[1] 0.9996425

my question is: I need to fit the data to an exponential distribution as well and compare the results, so I'm looking for a way to create the fits that returns comparable parameters. If there's better ways to find a power law fit, I'd be happy to try them.

Thank you

Answer 1

The poweRlaw package fits the power law and exponential distributions. You can install it in the usual way

install.package("poweRlaw")

Look at the first example in the first vignette.