DOC: mention that genlogistic is one of several types of generalized logistic distributions

[dschmitz89]

What does this implement/fix?

THere are several generalizations of the logistic distribution, genlogistic implements one of them. Acc. to Wikipedia, it is often to referred as type 1/skew logistic.. This is now mentioned in the notes section.

[mdhaber]

Please post or email a screenshot of [1] so that I can verify. Also, please cite the source of "It is also referred to as the skew-logistic distribution", even if it's Wikipedia.

[dschmitz89]

Please post or email a screenshot of [1] so that I can verify. Also, please cite the source of "It is also referred to as the skew-logistic distribution", even if it's Wikipedia.

Screenshot from Johnson:

[mdhaber]

Looks accurate. Hopefully docs build this time so we can see how they render.

[mdhaber]

Thanks @dschmitz89.

If you'd be willing to add a note that the loggamma distribution is actually the expgamma distribution, we could review and merge that quickly, too. I would also support adding the alias expgamma and deprecating loggamma.

I think loglaplace is also worth looking at. In quick tests, I was not able to confirm that either the logarithm or the exponential of a loglaplace random variable were distributed according to the laplace distribution.

[dschmitz89]

@mdhaber @j-bowhay : would it make sense to add the expgamma renaming to the depracation list #15765 ? I have never deprecated anything but could take a look.

[mdhaber]

It would need a note to the mailing list first, and I think that the PR with the deprecation would go in before adding to the tracker. The tracker reminds us to execute the removal of the deprecated behavior in two versions. (And a note in the existing documentation would also be ok with me, since it's less work.)