One problem saying:

Is $\sum_{n=1}^{\infty}1/(x^{2}+n^{2})$ uniform convergence?

So I solved it using Weierstrass M test since $\mathbb R $ is complete and $g_k\leq 1/k^2$ and the series consisting of $1/k^2$ is convergent by P-series test.

But what I want to know is the following problem:

Is $\sum_{n=1}^{\infty}x^{n}/n^{2}$ uniformly convergent on $x\in [0,1]$?

To solve it, I tried to use Cauchy Criterion.

So, I tried to find, for given $\epsilon$, certain $N$ for that condition.

For $k\geq N$


How can I find such $N$?

  • $\begingroup$ No, it cannot be convergent, as $\sum_{n=1}^\infty x^n/n^2$ has convergence radius $1$ $\endgroup$ – Ákos Somogyi Nov 16 '15 at 9:34
  • $\begingroup$ Your second series is convergent in $[-1,1]$ by the $M$-test. $\endgroup$ – Martín-Blas Pérez Pinilla Nov 16 '15 at 9:34
  • $\begingroup$ Ah! It's all my fault. problem gives me "Only $x\in[0,1]$" condition $\endgroup$ – Darae-Uri Nov 16 '15 at 9:36

If $n \geq 1$, then $$ x^{n}/n^{2} \leq 1/n^{2} $$ for all $x \in [0,1]$; by comparison test the uniform convergence follows.

  • $\begingroup$ Sorry for confusion. I fixed it. Thank you for your answer. $\endgroup$ – Darae-Uri Nov 16 '15 at 9:41
  • $\begingroup$ Then what value should I take as $N$? If I choose $N=1/\sqrt \epsilon$ then for $n\geq N$, $||g_k (x)+\cdots +g_{k+p} (x)||\leq||1/n^2 +\cdots +1/n^2 ||=p/\epsilon. $ $\endgroup$ – Darae-Uri Nov 16 '15 at 9:56
  • $\begingroup$ @Darae-Uri I was suggesting a quicker and to me more elegant way to see the uniform convergence; Cauchy's stuff can make things messy. :) $\endgroup$ – Megadeth Nov 16 '15 at 10:12
  • $\begingroup$ Ah! I see. Thank you for your comment. :D $\endgroup$ – Darae-Uri Nov 16 '15 at 10:13
  • $\begingroup$ @Darae-Uri Sorry for the careless version of my previous answer. $\endgroup$ – Megadeth Nov 16 '15 at 10:17

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.