Computation of sigma in ITU-R P.1853

[esqt]

Hello,

I might be wrong on this one, but it seems to me that your implementation of step A3 of ITU-R P.1853-1 is not exactly correct: it is using the inverse of the CDF of the normal distribution (ppf), instead of the Q function. This results in a negative sigma, which is by definition impossible.

ITU-Rpy/itur/models/itu1853.py

Line 140 in 1c7155c

Q = stats.norm.ppf((Pi / 100))

If I'm not mistaken, this should be easily fixed by adding a minus sign in front of stats.norm.ppf for steps A3 and C1.

Anyways, thanks for the excellent work.