I second the motion. I recently wrote a similar little fortran program
to calculate erf and think a set of functions for distribution manipulation
would be great (the equivalent of norminv, normdist, and normsdist
and thier gamma equivalents in excel, for example).
Since we mention free code, I have downloaded and looked at the
function code from gnumeric but never got around to using it.
There may be some easily modified but useful functions there, like the three
above.
Jan
Jeff Whitaker wrote:
> Dennis: The reason I needed erfc was to compute the CDF of a Gaussian. I
> think it might be more generally useful to include NCL functions that
> compute CDFs and PDFs of certain distributions (Gaussian and Gamma,
> especially). There's some free c-code to do this at
> http://www.stat.unipg.it/stat/dcdflib/.
>
> -Jeff
>
> --
> Jeffrey S. Whitaker Phone : (303)497-6313
> Meteorologist FAX : (303)497-6449
> NOAA/OAR/CDC R/CDC1 Email : jsw@cdc.noaa.gov
> 325 Broadway Web : www.cdc.noaa.gov/~jsw
> Boulder, CO, USA 80303-3328 Office : Skaggs Research Cntr 1D-124
-- Jan F. Dutton, PhD MBA 434-295-9989 President 801-457-0153 (efax) Weather Ventures, Ltd jfd@weatherventures.comP.O. Box 1851 Charlottesville, VA 22903 http://www.weatherventures.com
This archive was generated by hypermail 2b29 : Tue Apr 09 2002 - 08:40:32 MDT