I'd like your help with proving or refuting the following claim:

If for every $n \in \mathbb{N}$, $f_n:[a,b] \to \mathbb{R} $ is an increasing function, and if $f_n \to f$ pointwise in $[a,b]$ and $\sum \limits_{n=1}^{\infty}f_n(a)$ and $\sum \limits_{n=1}^{\infty}f_n(b)$ converge, then $f_n\to f$ uniformly in $[a,b]$.

I didn't come to any smart conclusion.

  • 3
    $\begingroup$ This is kind of a strange question, but if it's written correctly, then here's a hint: if $\sum_{n=1}^\infty f_n(a)$ converges, what can you say about $\lim_{n \to \infty} f_n(a)$? $\endgroup$ – Nate Eldredge Dec 13 '11 at 22:34
  • 3
    $\begingroup$ Here's two more hints: what can you say about $\lim_{n\rightarrow\infty }f_n(b)$? Then, what would $f_n$ increasing tell you? $\endgroup$ – David Mitra Dec 13 '11 at 22:56
  • $\begingroup$ It says that $\lim_{n \to infty}f_n(a)=0$, $\lim_{n \to infty}f_n(b)=0$, Does it tell me that $f_n=0$? $\endgroup$ – Jozef Dec 14 '11 at 7:40

The statement is true. In fact, $\{f_n\}$ converges uniformly to the 0 function on $[a,b]$:

Let $\epsilon>0$. Since $\sum\limits_{n=1}^\infty f_n(a)$ converges, $f_n(a)\rightarrow 0$. Thus, there is an $N_1$ such that $|f_n(a)|<\epsilon$ for all $n\ge N_1$. A similar argument gives us an $N_2$ such that $|f_n(b)|<\epsilon$ for all $n\ge N_2$.

Let $N=\max\{N_1,N_2\}$. Let $n\ge N$ and $x\in[a,b]$. Then, since $f_n$ is increasing: $$ -\epsilon< f_n(a) \le f_n(x) \le f_n(b)<\epsilon; $$ and so, $|f_n(x)|<\epsilon$ for all $n\ge N$ and $x\in[a,b]$.

Thus, $\{f_n\}$ converges uniformly to 0 on $[a,b]$.


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.