I really don't get why $$\sum_{j=0}^{\infty} \sum_{k=0}^j a_{k,j-k} =\sum_{j=0}^{\infty} \sum_{k=0}^{\infty} a_{k,j}$$

The only thing that makes sense to me is that if I write the terms I have

$$\sum_{j=0}^{\infty} \sum_{k=0}^j a_{k,j-k} =\sum_{j=0}^{\infty} (a_{0,j}+a_{1,j-1}+...+a_{j-1,1}+a_{j,0})$$

And then rearranging the terms I would eventually get the RHS on the first equation, but I have negative index so I'm not sure if this is rigorous or even correct.

Can someone explain me why the equality holds?

Thanks for your time.

  • 2
    $\begingroup$ Without a whiteboard it is hard to explain it, but you are essentially counting dots in the positive quadrant. There are different ways to count it; you can go row by row, or you can count downwards diagonals (which correspond to your two different sums). Also, I believe this equality does not necessarily hold; it will hold if the sequence a is absolutely convergent, or if they are all non-negative. $\endgroup$ – E-A Mar 13 at 1:02

Summing over $\mathbb Z^2$ diagonally gives the first series while summing vertically gives the second:


As mentioned in the comments, absolute convergence allows us to sum the infinite series in different orders, but for conditionally converging series they may not be equal.

| cite | improve this answer | |

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.