Full_Name: Jerry W. Lewis Version: 2.6.2 OS: Windows XP Professional Submission from: (NULL) (198.180.131.16) choose() returns incorrect values for all fractional arguments, regardless of sign. It returns 0 when both arguments are negative integers, which is not always correct (as in some formulations of the negative hypergeometric). Examples (correct answers from Maple's binomial function): choose( 4.75,2.75) # should be 8.90625 choose(-0.75,-0.5) # should be 1.6692536833481464 choose(-1.75,-0.5) # should be 0.55641789444938212 choose(-2.75,-1.5) # should be 0.15897654127125204 choose(-5, -7) # should be 15 The correct answers for first two can be verified easily since choose(n,k) should equal 1/beta(n-k+1,k+1)/(n+1) but does not for fractional arguments. The latter three are instances where beta incorrectly returns NaN, as noted in PR#10763
choose incorrect for fractional and some negative integer values (PR#10766)
2 messages · Jerry.Lewis at biogenidec.com, Ben Bolker
1 day later
<Jerry.Lewis <at> biogenidec.com> writes:
Full_Name: Jerry W. Lewis Version: 2.6.2 OS: Windows XP Professional Submission from: (NULL) (198.180.131.16) choose() returns incorrect values for all fractional arguments, regardless of sign. It returns 0 when both arguments are negative integers, which is not always correct (as in some formulations of the negative hypergeometric). Examples (correct answers from Maple's binomial function): choose( 4.75,2.75) # should be 8.90625
[snip] The help page does say that k is supposed to be an integer -- so I don't think this is a bug. Should the function either (a) test for k == (or close to ==) an integer or (b) be rewritten to work more generally? Ben