[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[igraph] Significance level of Kolmogorov-Smirnov test in power_law_fit(
From: |
Daniele Capocefalo |
Subject: |
[igraph] Significance level of Kolmogorov-Smirnov test in power_law_fit() output |
Date: |
Thu, 15 Jun 2017 11:30:38 +0200 |
User-agent: |
Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101 Thunderbird/52.1.1 |
Hello everyone,
I was trying to fit a power law on a dataset of ~15K nodes and ~155K edges.
I obtained the following output (from the summary of the PowerLawFit object:
'Fitted power-law distribution on discrete data\n\nExponent (alpha) =
3.073111
Cutoff (xmin) = 141.000000
Log-likelihood = -1589.823179
H0: data was drawn from the fitted distribution
KS test statistic = 0.024000
p-value = 0.997102
H0 could not be rejected at significance level 0.05'
so as you can see we are slightly above the critical point for
scale-freeness assumption.
My question is how to interpret the p-value: H0 in this case means that
the data are fit to a power law and therefore I can accept that alpha?
Thanks for your time,
Cheers,
D.
--
———
Daniele Capocefalo
Bioinformatics unit
Casa Sollievo della Sofferenza - Mendel
Viale Regina Margherita 261 - 00198 Roma IT
Tel: +39 06 44160526 - Fax: +39 06 44160548
E-mail: address@hidden
Web page: http://www.css-mendel.it/
Web page: http://bioinformatics.css-mendel.it
[Prev in Thread] |
Current Thread |
[Next in Thread] |
- [igraph] Significance level of Kolmogorov-Smirnov test in power_law_fit() output,
Daniele Capocefalo <=