Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

I was looking at the identity $\binom{n}{r} = \binom{n-1}{r-1} + \binom{n-1}{r}, 1 \leq r \leq n$, so in the case $r = n$ we have $\binom{n}{n} = \binom{n-1}{n-1} + \binom{n-1}{n}$ that is $1 = 1 + \binom{n-1}{n}$ thus $\binom{n-1}{n} = 0$, am I right?

share|improve this question
    
Removed the "(which also says that $(-1)! = 0$)", which was a silly mistake :$ –  Edgar Sánchez Jan 23 '13 at 5:48

2 Answers 2

up vote 6 down vote accepted

This is asking how many ways you can take $n$ items from $n-1$ items - there are none. So you are correct.

share|improve this answer
    
OK, but then we really do need $(-1)!$ defined as 0, right? –  Edgar Sánchez Jan 23 '13 at 5:18
    
Except for the part about $(-1)!=0$ --- I don't see where that comes from, and it's not the usual convention. –  Gerry Myerson Jan 23 '13 at 5:19
    
It would come from $\binom{n-1}{n} = \frac{(n-1)!}{n! (n-1-n)!}$ –  Edgar Sánchez Jan 23 '13 at 5:20
1  
That would put zero in the denominator. You really need $(-1)!=\infty$. And, indeed, the gamma function has poles at the non-positive integers. –  Thomas Andrews Jan 23 '13 at 5:29
2  
@EdgarSánchez But then you divide by $(-1)!$ ;) You don't want to divide by 0... BTW $\Gamma(-1)=\infty$ which suggests that the natural choice for $(-1)!=\infty$... –  N. S. Jan 23 '13 at 5:29

In my small treatize about "sums of like powers" (pg 14 ff) where I "re-invented" the integrals of the Bernoulli polynomials by a matrix-approach, I arrived at a formulation, where setting $ x=\binom{n}{n+1}$ to $ {1 \over (n+1)*(-1)!} \ne 0 $ was a meaningful thing, because it appeared in a formalism, where writing $ \frac 11b + \frac 12b + \frac13b + ... = (\frac 11 + \frac 12 + \frac 13 + ...)b = \frac {\zeta(1)}b = -1$ (where $b = \frac 1{(-1)!} $) allowed to arrive at the correct solution for the Bernoulli-/Zeta-polynomials. (We can argue, that $\lim _{\delta \to 0} { \zeta(1+\delta) \over \Gamma(\delta) } = 1$ and $\lim _{\delta \to 0} { \zeta(1-\delta) \over \Gamma(-\delta) } = -1$)

This is not a proof, that $ x=\binom{n}{n+1}$ need not always "be taken as zero", but it made me seriously thinking, how (and where else) such a setting actually makes sense.

share|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.