Friday, 12 June 2015

combinatorics - How to prove a modified version of the binomial theorem?




The binomial identity:
(x + y)^n = \sum_{k = 0}^{n}{n \choose k}x^{n - k}y^{k}



seems closely related to:
(x + y)_n = \sum_{k = 0}^{n}{n \choose k}(x_{n - k})(y_{k}) For
(x)_{n} = (x) * (x- 1) * \dots * (x- n + 1), (x)_0 = 1
I'm trying to prove the second identity, for all x \in \mathbb{C}, so the standard induction proof does not apply. For positive x, y, a simple combinatorial argument should work, but I can't generalize to all x \in \mathbb{C}.


Answer



As in this answer, in general we have

\frac{d^n}{dt^n} f(t)g(t) = \sum_{k=0}^n \binom{n}{k} \left( \frac{d^k}{dt^k} f(t) \right) \left( \frac{d^{n-k}}{dt^{n-k}} g(t) \right).
(The proof is by induction on n, using the product rule and Pascal's identity.)



Now put f(t) = t^y and g(t) = t^x. Then the two sides of the above equation are
\frac{d^n}{dt^n} t^y t^x = \frac{d^n}{dt^n} t^{x+y} = (x+y)_n t^{x+y-n},
and
\begin{align} \sum_{k=0}^n \binom{n}{k} \left( \frac{d^k}{dt^k} t^y \right) \left( \frac{d^{n-k}}{dt^{n-k}} t^x \right) &= \sum_{k=0}^n \binom{n}{k} (y)_k t^{y-n+k} \cdot (x)_{n-k} t^{x-k} \\ &= \sum_{k=0}^n \binom{n}{k} (x)_k (y)_{n-k} t^{x+y-n}. \end{align}
Putting t = 1 gives the desired result.






Note that if we divide both sides of your equation by n! then (recalling that \binom{m}{n} = (m)_n/n!) we end up with Vandermonde's identity
\binom{x+y}{n} = \sum_{k=0}^n \binom{x}{n-k} \binom{y}{k}.


No comments:

Post a Comment

real analysis - How to find lim_{hrightarrow 0}frac{sin(ha)}{h}

How to find \lim_{h\rightarrow 0}\frac{\sin(ha)}{h} without lhopital rule? I know when I use lhopital I easy get $$ \lim_{h\rightarrow 0}...