Degree of uncertainty

[Ilya Belalov]

Hi there,

I have submitted a paper to academic journal. I used powerlaw to fit my data. Among the reviewer comments there was the one, which is not clear how to respond to:

"No indication for the degree of uncertainty is given for the observations – and there clearly must be more uncertainty towards the right of each panel than the left. Similarly, the point-to-point variation in panel c suggests significant uncertainty across the entire panel. This uncertainty in the observations should be indicated so the reader can visually evaluate the strength of support for a heavy tail. Another place where quantifying uncertainty would be valuable is in comparing the fitted power-law exponents – what is the 95% confidence / credible interval for \alpha"

My data are distributed according to truncated power law. Would you advise me what I can do?

Thanks in advance,

Ilya

[Jeff Alstott]

Thanks for using powerlaw!

I've no suggestions for the issue of uncertainties-of-observations. As for the uncertainty in alpha, that is sigma, described in Clauset et al. and accessible in powerlaw as: fit.power_law.sigma

--
You received this message because you are subscribed to the Google Groups "powerlaw-general" group.
To unsubscribe from this group and stop receiving emails from it, send an email to powerlaw-gener...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/powerlaw-general/0b7c825f-dd1c-4ffd-8f40-cf98a6959c7bn%40googlegroups.com.

[Ilya Belalov]

The paper is out

https://journals.plos.org/ploscompbiol/article?id=10.1371/journal.pcbi.1008841&rev=2

[Jeff Alstott]

Congratulations!

To view this discussion on the web visit https://groups.google.com/d/msgid/powerlaw-general/25f5bf1a-be07-4cdf-ba10-9ca212936d7dn%40googlegroups.com.