Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I am trying to show that $\left\{ \left( 1-\dfrac {z} {\pi }\right) e^{\left( \dfrac {z} {\pi }\right) }\right\} \left\{ \left( 1+\dfrac {z} {\pi }\right) e^{\left( -\dfrac {z} {\pi }\right) }\right\} \left\{ \left( 1-\dfrac {z} {2\pi }\right) e^{\left( \dfrac {z} {2\pi }\right) }\right\} \left\{ \left( 1+\dfrac {z} {2\pi }\right) e^{\left( -\dfrac {z} {2\pi }\right) }\right\}\ldots $ Converges absolutely.

So the general term is of the form $\left( 1\pm \dfrac {z} {m\pi }\right) e^{\pm \dfrac {z} {m\pi }}$ I think it's the $e^{\pm \dfrac {z} {m\pi }}$ which is giving me grief. Any help would be much appreciated.

Edit: Is multiplying the $2m -1 $ and the $2m$ terms and canceling out the $e^{\pm \dfrac {z} {m\pi }}$ parts and $\left( 1^2\ - (\dfrac {z} {m\pi })^2\right)$ terms be allowed or would that run the risk of altering the value of the product ?

share|cite|improve this question
up vote 3 down vote accepted

This product is quite a famous one:

We have that the Gamma function defined as:

$$\Gamma(z) = \int\limits_0^\infty e^{-t} t^{z}\frac{dt}{t}$$

is convergent for $\Re(z)>0$

It is not too difficult to prove that, fo $n\in \mathbb{N}$,

$$\lim\limits_{n \to \infty} \frac{\Gamma({z+n})}{ n^{z-1}n!}=1$$

This with the functional equation of $\Gamma$ gives the identity:

$$\Gamma(z) = \lim\limits_{n \to \infty} \frac{n^{z} n!}{\prod\limits_{v=0}^{n} (z+v)}$$

Let's write

$$\tag{1} \frac{1}{\Gamma_n(z)} = \frac{\prod\limits_{v=0}^{n} (z+v)}{ n^{z} n!}$$

We know that

$$\lim\limits_{n \to \infty} \frac{1}{\Gamma_n(z)} =\frac{1}{\Gamma(z)} $$

We can write $(1)$ differently, as

$$ \frac{1}{\Gamma_n(z)} = \frac{z}{n^z}{\prod\limits_{v=1}^{n} \left(\frac{z}{v}+1 \right)}$$

Now, $${n^z} = {e^{z\log n}} = {e^{ - z\left( {1 + \frac{1}{2} + \frac{1}{3} + \cdots + \frac{1}{n} - \log n} \right)}}{e^{z\left( {1 + \frac{1}{2} + \frac{1}{3} + \cdots + \frac{1}{n}} \right)}}$$

so we can write

$$\frac{1}{{{\Gamma _n}(z)}} = z{e^{z\left( {1 + \frac{1}{2} + \frac{1}{3} + \cdots + \frac{1}{n} - \log n} \right)}}\prod\limits_{v = 1}^n {\left( {\frac{z}{v} + 1} \right)} {e^{ - \frac{z}{v}}}$$

Letting $n\to \infty$ gives:

$$\frac{1}{{\Gamma (z)}} = z{e^{z\gamma }}\prod\limits_{v = 1}^\infty {\left( {\frac{z}{v} + 1} \right)} {e^{ - \frac{z}{v}}}$$

where $\gamma\approx 0.577\dots$ is Euler's constant.

This means that

$$\frac{1}{{\Gamma ( - z)}} = - z{e^{ - z\gamma }}\prod\limits_{v = 1}^\infty {\left( {1 - \frac{z}{v}} \right)} {e^{\frac{z}{v}}}$$

or that

$$\eqalign{ & \frac{1}{{\Gamma ( - z)}}\frac{1}{{\Gamma (z)}} = - z{e^{ - z\gamma }}z{e^{z\gamma }}\prod\limits_{v = 1}^\infty {\left( {1 - \frac{z}{v}} \right)} \left( {1 + \frac{z}{v}} \right){e^{ - \frac{z}{v}}}{e^{\frac{z}{v}}} \cr & \frac{1}{{\Gamma ( - z)}}\frac{1}{{\Gamma (z)}} = - {z^2}\prod\limits_{v = 1}^\infty {\left( {1 - \frac{z}{v}} \right)} \left( {1 + \frac{z}{v}} \right) \cr} $$

But $$\frac{1}{{ - z\Gamma ( - z)}} = \frac{1}{{\Gamma (1 - z)}}$$ so

$$\frac{1}{{\Gamma (1 - z)}}\frac{1}{{\Gamma (z)}} = z\prod\limits_{v = 1}^\infty {\left( {1 - \frac{{{z^2}}}{{{v^2}}}} \right)} $$

Now put $z=\dfrac{s}{\pi}$, and your expression will appear. Note that the exponentials cancel out, and the infinite product converges everywhere.

share|cite|improve this answer

Your Answer


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.