MvNormalCDF.jl icon indicating copy to clipboard operation
MvNormalCDF.jl copied to clipboard

Error in description? Not _quasi_-Monte Carlo?

Open droodman opened this issue 1 year ago • 2 comments

The readme says "Quasi-Monte-Carlo numerical computation of multivariate normal probabilities." However, it looks to me like the "Quasi" bit is wrong. If the code uses random numbers, then it's just Monte Carlo simulation. If it uses number-theoretic, deterministic sequences such as Halton sequences---which maximize uniformity of coverage of the unit hypercube at the expense of independence of different draws---it is "quasi Monte Carlo."

droodman avatar Aug 11 '24 14:08 droodman

Hi! Yes, it use Richtmyer generator.

PharmCat avatar Aug 14 '24 22:08 PharmCat

Ah, I see. Thank you. I think it would be good to document that and explain the role of the rng argument.

droodman avatar Aug 15 '24 02:08 droodman