Message-ID: <437931186.230945.1294577869231.JavaMail.mail@webmail08>
Date: 2011-01-09T12:57:49Z
From: cgw at witthoft.com
Subject: how to calculate this natural logarithm
You could re-do part of your code to run with mpfr-class variables, and use
this function:
# mpfr choose(n,k)
rmpfac<-function(n,k,prec=50)
factorial(mpfr(n,prec))/factorial(mpfr(k,prec))/factorial(mpfr(n-k,prec))
Converting into and out of mpfr may not be worth it, but you will get all
the precision you want without any nasty "Inf" results
:-)
Carl