All you need is to take the inverse of the cdf on a uniform [0,1] random number. That's true for any continious distribution.
For the triangular one, it looks like G(p) from
http://www.brighton-webs.co.uk/distributions/triangular.asp
will work. Math not checked.
M.
_______________________________________________
gnumeric-list mailing list
[email protected]
http://mail.gnome.org/mailman/listinfo/gnumeric-list
