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.

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 have this function which is defined by infinite series


where $\{c_{n}\}$ is a sequence of nonzero real numbers such that $\sum \frac{1}{c_{n}}<\infty$.

My question is: Is $f$ bounded on $\mathbb R$? i.e. $|f(x)|<\infty$ for all $x\in \mathbb R$?

If not, can we make it bounded by assuming another condition on the sequence $\{c_{n}\}$?

share|cite|improve this question
You function will be discontinuous and unbounded in every $x=c_n$. – Pedro Tamaroff Mar 8 '12 at 2:39
As an example:$$\frac{1}{z} + \sum\limits_{k = 1}^\infty {{{\left( { - 1} \right)}^k}\left( {\frac{1}{{z + k}} + \frac{1}{{z - k}}} \right)} = \frac{\pi }{{\sin \pi z}}$$ But at every $z \in \mathbb{N}$ we're in trouble. – Pedro Tamaroff Mar 8 '12 at 3:19
This is not quite an example of the specific sum Terra M is considering. – Robert Israel Mar 8 '12 at 3:47
@Robert I just wanted to provide a concrete example of a similar looking function, but, yes, you're right. – Pedro Tamaroff Mar 8 '12 at 4:15

As Peter observed, the limit function will not be bounded.

Indeed, one such example is

$$\sum_{n=1}^{\infty} \frac{1}{x^2-n^2} = \frac{\pi x \cot(\pi x) - 1}{2x^2},$$

where the sum converges uniformly on compact subsets of $\mathbb{C}\setminus\mathbb{Z}$. The limit function has a pole at every integer.

share|cite|improve this answer
That sum is not of the form given in the original question. – Gerry Myerson Mar 8 '12 at 5:13
@GerryMyerson, just replace $x$ with $\sqrt{x}$ and consider $x > 0$. – Antonio Vargas Mar 8 '12 at 5:18
On second thought, you don't even need to restrict yourself to positive $x$ after making the substitution $x \mapsto \sqrt{x}$; after making a suitable branch cut, the function is real-valued for all real $x \neq 0,1,2,\ldots$ – Antonio Vargas Mar 8 '12 at 5:33

By the way, convergence of $\sum_{n=1}^\infty \frac{1}{c_n}$ doesn't imply that $\sum_{n=1}^\infty \frac{1}{x-c_n}$ converges for $x \notin \{c_n: n=1\ldots\infty\}$. For example, take $c_{2n} = d_n$ and $c_{2n+1}=-d_n$ where $d_n$ is any sequence of positive numbers with $d_n \to \infty$. Then $\sum_{n=1}^\infty \frac{1}{c_n} = 0$. But since $\frac{1}{x-d_n} + \frac{1}{x+d_n} = \frac{2x}{x^2 - d_n^2}$, $\sum_{n=1}^\infty \frac{1}{x-c_n}$ diverges whenever $x \ne 0$ and $\sum_{n=1}^\infty 1/d_n^2$ diverges.

On the other hand, if $\sum_{n=1}^\infty \frac{1}{c_n}$ converges absolutely, $\sum_{n=1}^\infty \frac{1}{x-c_n}$ will also converge absolutely as long as $x \notin \{c_n: n = 1 \ldots \infty\}$, because $\frac{1}{|x - c_n|} \le \frac{2}{|c_n|}$ when $n$ is large enough that $|c_n| > 2 |x|$. The sum will then be a meromorphic function of $x$ with a pole at each $c_n$.

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.