Dear S-plus users!
Does anybody have an idea of code (algorithm) in S or C
to calculate bivariate normal integral ("F(h,k,ro)") having the joint
stand. bivariate normal density? My problem is to achieve a very high
precision
and to find algorithm that runs fast, because I need to calculate
F(h,k,ro) for huge vectors h,k, ro and to make this many times.
I tried some different approximations as posted in "Distributions in
Statistics", Johnson, Kotz . They are "fast" but not enough accurate.
All other methods such as numeric integration using Simpson's rule and
recursion are very slowly (even when the code is written on C).
Thanks for any contribution.
Regards, Viki.
-----------------------------------------------------------------------
This message was distributed by s-news@wubios.wustl.edu. To unsubscribe
send e-mail to s-news-request@wubios.wustl.edu with the BODY of the
message: unsubscribe s-news
|