Suggestion: implement Q-Q plots

[dsaxton]

I see that there was some discussion of this years ago (PR #1501), but was wondering if there are any plans of implementing these in the future.

One thought would be to have qqplot be a Series method which optionally takes another Series as an argument. If this argument is missing then qqplot could fall back on the quantiles from some theoretical distribution (default Gaussian, say), perhaps imported from statsmodels. Even without statsmodels integration a purely empirical method could be quite useful for providing a very fast way of comparing two distributions.

[gfyoung]

@dsaxton : Thanks for reporting this! Do you mind referencing the issue where you saw the original discussion about this?

cc @TomAugspurger

[TomAugspurger]

Since statsmodels already has Q-Q plots, I think further development should happen there.

…

On Sun, Aug 19, 2018 at 3:10 AM gfyoung ***@***.***> wrote: @dsaxton <https://github.com/dsaxton> : Thanks for reporting this! Do you mind referencing the original issue where you saw the original discussion? cc @TomAugspurger <https://github.com/TomAugspurger> — You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub <#22393 (comment)>, or mute the thread <https://github.com/notifications/unsubscribe-auth/ABQHIgM_O4_1_FOEfuLgn-vxrYHJ7y9yks5uSR18gaJpZM4WA1o_> .

[dsaxton]

That seems fair, closing the issue.