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

Which of the following statements are true?

  1. Let $\{a_{mn}\}, m ,n \in \mathbb{N}; $be an arbitrary double sequence of real numbers. Then $$\sum^ \infty_{m=1} \sum^ \infty_{n=1} a^3_{mn} = \sum^ \infty_{n=1} \sum^ \infty_{m=1} a^3_{mn} $$
  2. Let $\{a_{mn}\}, m ,n \in\mathbb{N}; $be an arbitrary double sequence of real numbers. Then $$\sum^ \infty_{m=1} \sum^ \infty_{n=1} a^2_{mn} = \sum^ \infty_{n=1} \sum^ \infty_{m=1} a^2_{mn} $$

  3. Let $\{a_{mn}\}, m ,n \in \mathbb{N}; $be an arbitrary double sequence of real numbers such that $|a_{mn} |\leq \sqrt{\frac{m}{n}} $ Then $$\sum^ \infty_{m=1} \sum^ \infty_{n=1} \frac{a_{mn}}{m^2n} = \sum^ \infty_{n=1} \sum^ \infty_{m=1} \frac{a_{mn}}{m^2n} $$

share|cite|improve this question
the last one seems to converge absolutely – Santosh Linkha Jan 10 '13 at 17:40
If the terms are positive (your second question), see here:… – Martin Sleziak Jan 23 '14 at 13:58


  1. Let $a_{nn}=n$ and $a_{n+1,n}=-n$ for $n\in\Bbb Z^+$, and let $a_{mn}=0$ in all other cases.

  2. This is true, assuming that you allow $\infty$ as a sum; how you prove it will depend on what you already know.

  3. $\dfrac{\sqrt{m/n}}{m^2n}=\dfrac1{\sqrt{m^3n^3}}=\dfrac1{m^{3/2}n^{3/2}}$, so for fixed $m$ the series $\displaystyle\sum_{n\ge 1}\frac{a_{mn}}{m^2n}$ is absolutely convergent, and for fixed $n$ the series $\displaystyle\sum_{m\ge 1}\frac{a_{mn}}{m^2n}$ is absolutely convergent.

share|cite|improve this answer
still not getting... – user45799 Jan 10 '13 at 17:43
@Prasanta: Which part? – Brian M. Scott Jan 10 '13 at 17:45
both 1 and 2... – user45799 Jan 10 '13 at 17:47
@Prasanta: For (1) just evaluate the two sums for the double series that I suggested. For (2) and (3) you need to know that equality holds if the double series is absolutely convergent. – Brian M. Scott Jan 10 '13 at 17:50
Maybe you intended a different sequence for 1. With your choice both sums equal $0$. Am I right? – P.. Jan 10 '13 at 17:54

Use the Fubini–Tonelli theorem. We have that $a_{mn}\leq\sqrt{\frac{m}{n}}= m^{\frac{1}{2}}\cdot n^{-\frac{1}{2}}$ implies $\frac{a_{mn}}{m^{\frac{1}{2}}\cdot n^{-\frac{1}{2}}}$ and
$$ \frac{a_{mn}}{m^{2}\cdot n^{1}} = \frac{a_{mn}}{ m^{\frac{1}{2}}\cdot n^{-\frac{1}{2}}} \cdot \frac{ m^{\frac{1}{2}}\cdot n^{-\frac{1}{2}}}{m^2n^1} \leq \dfrac{1}{m^{\dfrac{3}{2}}}\cdot\dfrac{1}{n^{\dfrac{3}{2}}} $$

Note that $\displaystyle\sum_{n=1}^{\infty}\frac{1}{ n^{3/2} }< \infty $ and $\displaystyle\sum_{m=1}^{\infty}\frac{1}{ m^{3/2} }< \infty $ implies $\displaystyle\sum_{n=1}^{\infty}\frac{1}{m^{3/2} n^{3/2} }< \infty $ and $\displaystyle\sum_{m=1}^{\infty}\frac{1}{ m^{3/2}n^{3/2} }< \infty $ for all $m,n\in\mathbb{N}^*$.

Set the measures in $\mathbb{N}$: $$ \mu(\{m\})=\frac{1}{m^{2}} \mbox{ and } \nu(\{n \})=\frac{1}{n}. $$ The above observations it follows that the hypotheses of the theorem Fubini-Tonelli theorem are satisfied. Therefore nostemos the following. \begin{align} \sum^ \infty_{m=1} \sum^ \infty_{n=1} \frac{a_{mn}}{m^2n}= & \int_{\mathbb{N}}\int_{\mathbb{N}} a(m,n) d\mu(m) d\nu(n) \\ = & \int_{\mathbb{N}}\int_{\mathbb{N}} a(m,n) d\nu(n) d\mu(m) \\ = & \sum^ \infty_{n=1} \sum^ \infty_{m=1} \frac{a_{mn}}{m^2n} \end{align}

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.