Erros in alpha calculation

[XauTrop]

When using your power law implementation with this data:
[0.0018, 0.0076, 0.0077, .... , 277.89, 546.74]
I got this error:
/usr/local/lib/python3.4/dist-packages/plfit/plfit.py:1012: RuntimeWarning: invalid value encountered in log
alpha = 1.0 + float(nn) * (sum(log(xx/(float(xmin)-0.5))))**-1

This is because your implementation is not working with values below 1?

[keflavich]

That's just a warning, but presumably you got alpha=nan? There shouldn't be a problem with xmin<1 generally, but it looks like there is a way to get negatives here... hm.

[XauTrop]

IThanks for the answer. I managed multipliying all the data to get them >1 and of course divide again the xmin. I'am curious if there is a way to get various xmin from the module, like a list of the maximum curvature change in the data, not only the maximum curvature. My data looks like that, and the maximum curvature is not the desired xmin. Thanks in advance.

[keflavich]

You might want https://github.com/keflavich/plfit/blob/master/plfit/plfit.py#L474 , alphavsks, and https://github.com/keflavich/plfit/blob/master/plfit/plfit.py#L455, xminvsks, to check the best-fit statistic vs. xmin value