I am wondering how to determine the number of observations that fall within the tail of a distribution. I am reading a paper and the authors use the assumption that 50 observations need to fall into the right tail of a power law distribution. I have a degree distribution from a network and now I would like to test if 50 observations fall into the tail, however, the tail is not defined in the paper as far as I know.

I would be grateful for some guidance.

| cite | improve this question | | | | |
  • $\begingroup$ There is no such hard and fast rule. The authors likely want 50 observations in the tail to have some assurance that the tail is a power function. The more there are, the stronger the assurance. However, this is also a vapid argument. One is better off fitting the entire data set with a distribution that becomes a power law in the tail, which is a much stronger argument. $\endgroup$ – Carl Mar 5 '18 at 11:35
  • $\begingroup$ @Carl I understand. But does any way come to mind? Because at this stage I need to try to replicate the results using their methodology and I do not exactly know how to do so. $\endgroup$ – abu Mar 5 '18 at 12:01
  • $\begingroup$ Will have to read paper closely to give advice. That may take a few days. In the meantime, I need to know more about the data you are are of a mind to collect, or have collected. $\endgroup$ – Carl Mar 5 '18 at 12:05

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.