Skip to main content
Logo image

A First Course in Complex Analysis

Section 10.2 Binomial Coefficients

The binomial coefficient \(\binom n k = \frac{ 1 }{ k! } n(n-1)(n-2) \cdots (n-k+1)\) is a natural candidate for being explored analytically, as the binomial theorem
\begin{equation*} (x+y)^n \ = \ \sum_{k=0}^n \binom n k x^k \, y^{n-k} \end{equation*}
(for \(x,y \in \C\) and \(n \in \Z_{ \ge 0 }\)) tells us that \(\binom n k\) is the coefficient of \(z^k\) in \((z+1)^n\text{.}\) You will derive two sample identities in the course of the exercises below.
  1. Convince yourself that
    \begin{equation*} \binom{n}{k} \ = \ \frac 1 {2 \pi i} \int_\gamma \frac{ (z+1)^{n} }{ z^{k+1} } \, \diff{z} \end{equation*}
    where \(\gamma\) is any simple closed piecewise smooth path such that 0 is inside \(\gamma\text{.}\)
  2. Derive a recurrence relation for binomial coefficients from the fact that \(\frac 1 z + 1 = \frac{ z+1 }{ z }\text{.}\) (Hint: Multiply both sides by \(\frac{ (z+1)^n }{ z^k }\text{.}\))
  3. Now suppose \(x \in \R\) with \(|x|\lt 1/4\text{.}\) Find a simple closed path \(\gamma\) surrounding the origin such that
    \begin{equation*} \s \left( \frac{ (z+1)^2 }{ z } \, x \right)^k \end{equation*}
    converges uniformly on \(\gamma\) as a function of \(z\text{.}\) Evaluate this sum.
  4. Keeping \(x\) and \(\gg\) from 3, convince yourself that
    \begin{equation*} \s \binom{2k}{k} x^k \ = \ \frac 1 {2 \pi i} \s \int_\gamma \frac{ (z+1)^{2k} }{ z^{k+1} } \, x^k \, \diff{z} \, \text{,} \end{equation*}
    Use (3) to interchange summation and integral, and use the Residue Theorem 9.2.2 to evaluate the integral, giving an identity for \(\s \binom{2k}{k} x^k\text{.}\)