Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

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 am learning about uniform convergence of function series. I wanted to ask whether I answered this exercise correctly:

Let $$f_n=\frac{\sqrt{n}\cdot \sin(x)}{nx^{2}+2}$$

  1. Does $f_n$ uniformly converge in $[0,\infty)$
  2. Does $f_n$ uniformly converge in $[-\pi,\pi]$?
  3. Does $\displaystyle\int_{\pi}^{2\pi}f_n(x)dx \rightarrow 0$?


$f_n$ is point-convergent to $f(x)=0$ in R. Put $x_n=\frac{1}{\sqrt{n}}$, then $|f_n(x_n)-f(x_n)|\rightarrow \frac{1}{3}$, and therefore the answer to 1. and 2. is that $f_n$ is not uniformly convergent in those intervals.

For 3., the answer is yes since $f_n(x)\rightarrow f(x)=0$.

What is odd about this solution to me is that you can answer 1. and 2. seemingly in the same way. So, is this correct?


share|cite|improve this question
Thank you everyone for the answers! I wish I could've chosen them all :). – ro44 Jan 27 '12 at 15:13
up vote 1 down vote accepted

Your argument in (3) is wrong: Consider a sequence of piecewise linear functions $(f_n)$ where the graph of $f_n(x)$ is a straight line through $(0,0)$ and $(\frac1{2n},2n)$ for $0 \leq x \leq \frac1{2n}$ and a straight line through $(\frac1{2n},2n)$ and $(\frac1n,0)$ for $\frac1{2n} \leq x \leq \frac1n$ and $0$ everywhere else.

Then we have for all $x \in [0,1]$ that $f_n(x) = 0$ eventually. In particular $f_n \to 0$ pointwise, but: $$ \lim_{n\to \infty} \int_0^1 f_n(x)\,dx = 1 \neq 0 = \int_0^1 \lim_{n\to \infty} f_n(x)\,dx. $$

This is a standard example of why pointwise convergence and integrals is a delicate question.

share|cite|improve this answer
Thank you for your input! I guess the answer in (3) is dependent on whether $f_n$ is uniformly convergent in $[pi,2pi]$, which I think it is since $|f_n-0|<\frac{1}{\sqrt{n}}$ in that interval. So the answer is still yes for different reasons. – ro44 Jan 27 '12 at 15:08
@ro44 Kahen's point is that you only showed pointwise convergence. It is true that if $f_n$ converges to $0$ uniformly on $[a,b]$ then $\int_a^b f_n(x)dx$ converges to $0$. In other words, your argument above is wrong, but the "yes, it does converge to zero" answer is correct. – Thomas Andrews Jan 27 '12 at 15:45

Your solution for 1) and 2) is correct. Your answer for part 3) is correct, but the reasoning is flawed. Pointwise convergence does not imply convergence of the integrals; uniform convergence, however, does. That is, if $\{ g_n\}$ converges uniformly to $g$ on $[a,b]$ and if each $g_n$ is integrable on $[a,b]$, then $g$ is integrable on $[a,b]$ and $\int_a^b g_n\rightarrow\int_a^b g$.

You can show that your $f_n$ converge uniformly to 0 on $[\pi, 2\pi]$ (which leads me to believe that part 2) should have asked if the convergence was uniform on $[\pi,2\pi]$).

share|cite|improve this answer

For 3. pointwise convergence isn't strong enough (as other users have already answered). The good news is that your sequence of functions converges uniformly in $[a, +\infty], \forall \ a>0$, and therefore also in $[\pi, 2\pi]$. To prove this, simply reason as follows:

$$\sup_{[a,+\infty]}{|f_n|}=\sup_{[a, +\infty]}{\frac{\sqrt{n}\cdot \sin(x)}{nx^{2}+2}} \leq \frac{\sqrt{n}}{na^{2}+2} \rightarrow 0$$

for $n \rightarrow\infty$. Your problem with uniform convergence in $\mathbb{R}$, as your solution to 1. shows, is located "close to $0$"; if you move away from $0$ this problem disappears.

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.