By absolute ECDF, I am referring to an ECDF that shows the absolute number of samples above some value as opposed to the fraction of such samples. For example, the bottom plot below displays such an absolute ECDF, while the top plot is a regular ECDF of the same data.

top: ECDF, bottom: absolute ECDF

When deciding about features for a possible ECDF implementation for Matplotlib, the question arose whether it makes sense to also implement such absolute ECDFs. To this end, I wanted to search for actual uses or teaching material on those, but failed to find anything. There are two conceivable reasons for this:

  • I am not using the right term.
  • Nobody uses this kind of visualisation.

Thus I am asking: Is there any established term for what I call absolute ECDF?

  • 1
    $\begingroup$ Your "absolute ECDF" reminds me of Pareto charts, which however differ in that they order the contributing factors in decreasing order of contributions (yielding a convex cumulative line), which is why I am not posting this as an answer. Alternatively, people may use either counts/frequencies or cumulative probabilities and simply not see the point in using a different name in each case. Compare the freq parameter to hist() in R. $\endgroup$ Feb 27 '20 at 10:36
  • 2
    $\begingroup$ Google "cumulative frequency chart." $\endgroup$
    – whuber
    Feb 27 '20 at 13:55

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.