MathOverflow is a question and answer site for professional mathematicians. It's 100% free, no registration required.

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

The function $f(x) = e^{2\pi ix}$ on the domain $\mathbb{R}/\mathbb{Z}$ has the property that, for every $n > 1$ and every $x$, $\displaystyle \sum_{i = 0}^{n-1} f(x + \frac{i}{n}) = 0$.

Other such functions can be found by simply postcomposing a linear function to this example (thus, for example, $x \mapsto \cos(2\pi x) : \mathbb{R}/\mathbb{Z} \to \mathbb{R}$ also has this property).

Beyond those, are there any other "natural" examples of functions with this property (on the same domain $\mathbb{R}/\mathbb{Z}$, but with any codomain)?

[Of course, one can freely construct the codomain as a monoid with generators $\mathbb{R}/\mathbb{Z}$ and all the necessary relations; slightly less trivially, one could take the codomain to be a ring, the map to be exponential, and $f(\frac{1}{n}) - 1$ to be invertible for each prime $n$, in the same freely constructed fashion. I can't quite put my finger on why, but I don't want to count these as "natural" examples (probably because the condition for each (or essentially each) distinct $n$ is handled separately, instead of flowing all at once from some underlying property)]

share|cite|improve this question

Every continuous integrable function satisfying your hypothesis is a linear combination of $\sin(2\pi x)$ and $\cos (2\pi x)$. The key fact is that if $f:\mathbb{R}/\mathbb{Z}\rightarrow \mathbb{C}$ is continuous and integrable then it is uniquely determined by its Fourier coefficients. If the Fourier series expansion of $f$ is $$f(x)=\sum_{n\in\mathbb{Z}}c(n)e(nx),$$ where $e(x)=\exp (2\pi i x)$, then for $n>1$ the Fourier expansion of $$g_n(x):=\sum_{a=1}^nf\left(x+\frac{a}{n}\right)$$ is $$g_n(x)=\sum_{m\in\mathbb{Z}}c(m)\left(\sum_{a=1}^ne(ma/n)\right)e(mx)$$ $$=n\sum_{m\in\mathbb{Z}}c(mn)e(mnx).$$ If $g_n(x)=0$ for all $n>1$ and for all $x$ then we must have that $c(mn)=0$ for all $m\in\mathbb{Z}$ and for all $n>1$, and this proves the result.

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.