variance_mc

[kcarloni]

Hi, thanks for your work on this package!

Is the calculation of the mc variance correct? (Line 137)
variance_mc = (sum(Jsf.^2)/ncalls - integral_mc^2) / (ncalls - 1) + eps()

I think it should be
variance_mc = (sum(Jsf.^2)/ncalls - integral_mc^2) * ncalls/(ncalls - 1) + eps()

to implement Bessel's correction properly.

[ranjanan]

Check Equation 6 from https://arxiv.org/pdf/2009.05112.pdf?

[kcarloni]

Hi, thanks for your reply. I agree the implementation matches the paper -- I think that equation 6 might itself be wrong, but it does make sense to stay consistent with the published version