ENH: stats: Non-parametric quantile test

Hello everyone,

I would like to propose the addition of a non-parametric test for 
quantiles.

Essentially, it as the median analog of a one-sample t-test: just as 
`ttest_1samp` tests whether the population mean is a specified value and 
the result object has a method for computing the confidence interval of 
the population mean, this new function (simply named `quantile_test`) 
tests whether the median is a specified value and the result object has 
a method for computing the confidence interval of the population median.

Differences are:
- it works for any quantile (not just the median)
- it is non-parametric.

Huge thanks to Matt who did all the heavy lifting to help putting this 
PR in shape!

If you're interested, please join the discussion at: 
https://github.com/scipy/scipy/pull/12680

Cheers,

-- 
Romain Jacob
Postdoctoral Researcher
ETH Zurich - Networked Systems Group (NSG)
www.romainjacob.net <https://www.romainjacob.net/>
@RJacobPartner <https://twitter.com/RJacobPartner>
Gloriastrasse 35, ETZ G81
8092 Zurich
+41 7 68 16 88 22

_______________________________________________
SciPy-Dev mailing list -- [email protected]
To unsubscribe send an email to [email protected]
https://mail.python.org/mailman3/lists/scipy-dev.python.org/
Member address: [email protected]