Sorry for any cross-posting.
Thank you. I do not need any more replies.
Some have asked for a summary.
Summary of Responses
I have received many e-mails (at least 30). Thank you
for your help. It was interesting to see the different
values obtained from 5.58 to 5.645 to 5.612 to 5.066
to 5.19. But the majority give 5.326724 as the answer
(obtained from SAS, R, SPSS and SPlus). This is good
since it agrees with the output from my algorithm. I
get the same value as SAS up to 8 d.p. as
5.32672389
It is a good check of the accuracy of software using
the very tail-end values of a distribution.
Thank you.
Sincerley,
Paul Johnson
http://www.biostatsoftware.com
|