-----Original Message----- From: boost-users-bounces@lists.boost.org [mailto:boost-users- bounces@lists.boost.org] On Behalf Of Belcourt, Kenneth Sent: Monday, September 06, 2010 4:23 PM To: boost-users@lists.boost.org Subject: Re: [Boost-users] looking for some help to find some functionality
On Sep 4, 2010, at 11:03 AM, Paul A. Bristow wrote:
-----Original Message----- From: boost-users-bounces@lists.boost.org [mailto:boost-users- bounces@lists.boost.org] On Behalf Of Steven Watanabe Sent: Friday, September 03, 2010 10:00 PM To: boost-users@lists.boost.org Subject: Re: [Boost-users] looking for some help to find some
3. Availability of Normal Density Function, Cumulative Normal Density Function, and Inverse Cumulative Normal Density.
http://www.boost.org/libs/math/doc/sf_and_dist/html/math_toolkit/dist/ dist_
ref/dists/normal_dist.html
normal (Gaussian) Yes -
But we don't - yet- have the *inverse* normal distribution - but I am just adding some more distributions (inverse gamma and inverse chi squared) and I might add this one too if there is demand. It's not trivial, but seems to have been studied and there are some published implementations that could probably be adapted quite easily for the Boost.Math multi-precision format.
Hi Paul,
I coded up an inverse gaussian years ago for use with an LHS (latin hypercube sampling) program. I attach the files though regrettably I didn't make a note of where the coefficients came from (probably from the original LHS Fortran 77 application).
Don't know if they're any help but thought I'd send them along.
Thanks for these. http://home.online.no/~pjacklam/notes/invnorm/impl/lea/lea.c also looks promising for inverse gaussian. Paul