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 want to prove that the product $$\prod_{n=1}^\infty \left(1+\frac{z}{n} \right) \mathrm{e}^{-\frac{z}{n}}$$ converges absolutely, and uniformly on compact subsets of $\mathbb C$:


My book (Ahlfors) defines the absolute convergence of the product $\prod a_n$ by the absolute convergence of the series $\sum \text{Log } a_n$, where only a finite number of terms can be zero, and they must be removed from the sequence. ($\text{Log}$ denotes the principal branch of the logarithm, where the argument is in $(-\pi ,\pi]$.)

Thus, in order to prove absolute convergence, I fix $z \in \mathbb C$, and look at the series whose terms are $\text{Log } \left[ \left(1+\frac{z}{n} \right) \mathrm{e}^{-\frac{z}{n}} \right]$. It is well known that if $\text{Arg }z_1,\text{Arg }z_2$ both lie in $(-\frac{\pi}{2},\frac{\pi}{2})$, then $\text{Log } (z_1z_2)= \text{Log } z_1+\text{Log } z_2$. For large enough $n$ both factors $z_1=\left( 1+\frac{z}{n} \right),z_2=\mathrm{e}^{-\frac{z}{n}}$ satisfy this, so $\text{Log } \left[ \left(1+\frac{z}{n} \right) \mathrm{e}^{-\frac{z}{n}} \right]=\text{Log } \left(1+\frac{z}{n}\right)+\text{Log } \mathrm{e}^{-\frac{z}{n}}$. It is also possible to prove that $\text{Log } \mathrm{e}^{-\frac{z}{n}}=-\frac{z}{n}$ for large enough $n$. The problem thus reduces into the convergence of $$\sum \left\lvert \text{Log } \left(1+\frac{z}{n} \right)-\frac{z}{n} \right\rvert $$

Taylor's theorem states that $$\text{Log }(1+w)=w+w^2 g(z) .$$ I've estimated the remainder and found that for $|w|<\frac{1}{2}$, $$| g(w)| \leq 2(\ln 2+\pi), $$ therefore, for large enough $n$ (so that $|\frac{z}{n}| < \frac{1}{2}$), the terms of the last series are dominated by $2(\ln2+\pi)\frac{|z|^2}{n^2}$, and $\sum 2(\ln2+\pi)\frac{|z|^2}{n^2}$ converges.

I deliberately left the $\Sigma$'s unindexed, since $z$ might be a zero of one of the factors. In that case only a tail of the series is taken into account.


The proof of uniform convergence on compact sets follows the same lines. Let $K \subset \mathbb C $ be a compact set, as such it lies in some ball $|z| \leq M$. Taking large enough $n$, we find similarly, that the series $\sum \left\lvert \text{Log } \left[ \left(1+\frac{z}{n} \right) \mathrm{e}^{-\frac{z}{n}} \right] \right\rvert$ is dominated by $\sum 2(\ln2+\pi)\frac{M^2}{n^2}$ which converges. The Weierstrass M-test says that the series is uniformly convergent, and hence, so is the product.

I'd love to hear any thoughts/remarks about these proofs.

share|improve this question
    
I find it OK. The general case is considered in W. K. Hayman, Meromorphic functions, Claredon Press, Oxford, 1964, paragraph 1.10 and here, paragraphs 2.3, 2.4. –  user64494 Aug 8 '13 at 18:48

1 Answer 1

up vote 1 down vote accepted

Looks fine. It's natural to take logarithm for a product, but I think there should be a more direct way to define absolute convergence of a product besides literally defining it in terms of the sum of logarithm.

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.