Computational issues with @cnorm

[Novi]

Hello everybody!
Once again a very geeky question for you, I hope there is somebody out there who knows the answer...
I searched the manual many and many times but I couldn't find out how does Eviews compute the @cnorm function: clearly it is a complex one, involving the computation of an integral and consequent approximation, nevertheless no where it is explained what's behind the command @cnorm.

Does any of you know more about it?

Thanks so much

[EViews Glenn]

It's based on near-minimax approximations to the error function, since the cumulative normal can be written as a function of the former. "Rational Chebyshev approximations for the error function" W. J. Cody, Math. Comp., 1969, PP. 631-638.

[Novi]

Thanks a lot Glenn :)