[SciPy-Dev] stats.hypsecant2

[Andrew Nelson]

Does anyone have any objections to adding a secant-squared distribution to
stats?
The CDF of this distribution (hyperbolic tangent) is frequently used for
modelling scattering data (see appendix A in
https://doi.org/10.1107/S1600576717013632, eqns 25 and 26).

CDF = 0.5 * {1 + tanh(pi * (x - loc) / 2 / sqrt(3) / scale)}
PDF = (pi / 4 / sqrt(3) / scale) * (cosh(pi * (x - loc) / 2 / sqrt(3) /
scale))**-2

I'm guessing the name should be hypsecant2

Andrew.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/scipy-dev/attachments/20180202/79594548/attachment.html>