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

I was looking at this question, and trying to come up with a counterexample. After thinking about it, I thought the following might be true:

Claim: let $\{f_n\}$ be a sequence of continuous, monotonically increasing functions on a compact $K \subset \Bbb R$. That is, each $f_n : K \to \mathbb{R}$ satisfies $x < y \implies f(x) \leq f(y)$.

Suppose $f_n \to f$ pointwise, where $f$ is a continuous function. Then the family $\{f_n\}$ is equicontinuous.

Is it true? If so, I'd like to prove it, and if not, I'd like a counterexample.

My thoughts so far: I think it's true, but perhaps this is due to a lack of imagination. It's clear that $f$ must be monotonically increasing. Also, $f_n - f$ is a sequence of continuous functions that converges to $0$ pointwise.

I was also thinking of perhaps starting with a lack of continuity and proving the existence of a jump discontinuity in $f$, though it's not clear this would work.

Any input is appreciated!

EDIT: This is much healthier's answer in the original question suggests a certain working $\epsilon$-$\delta$ proof, so if anyone wants to work through that I'd appreciate it. Alternate approaches would be even more interesting to me.

share|cite|improve this question
You say f is increasing. Is it the sequence that is increasing or the functions themselves? – mathematician Jul 2 '14 at 19:18
The functions themselves. I'll clarify. – Omnomnomnom Jul 2 '14 at 19:20
That is, we do not (necessarily) have $f_n(x) \leq f_{n+1}(x)$ for all $x \in K$. – Omnomnomnom Jul 2 '14 at 19:22

Such a family is equicontinuous, and hence uniformly equicontinuous, since $K$ is compact.

Fix $x\in K$ arbitrarily, and let $\varepsilon > 0$. Since $f$ is continuous and monotonically increasing (not necessarily strictly), there is a $\delta > 0$ such that for every $y \in [x-\delta,x+\delta] \cap K$ we have $f(x) - \varepsilon/4 \leqslant f(y) \leqslant f(x) + \varepsilon/4$.

Let $a = \min [x-\delta,x]\cap K$ and $b = \max [x,x+\delta]\cap K$. Since $f_n \to f$ pointwise, there is an $n_0$ such that $\lvert f_n(a) - f(a)\rvert \leqslant \varepsilon/4$ and $\lvert f_n(b) - f(b)\rvert \leqslant \varepsilon/4$ for all $n \geqslant n_0$.

For $n \geqslant n_0$, we then have $\lvert f_n(y) - f_n(x)\rvert \leqslant f_n(b) - f_n(a) \leqslant (f(x)+2\varepsilon/4) - (f(x) - 2\varepsilon/4) = \varepsilon$ for all $y \in [x-\delta,x+\delta]\cap K$.

For each of the finitely many $n < n_0$, there is a $\delta_n > 0$ such that $\lvert f_n(y) - f_n(x)\rvert \leqslant \varepsilon$ for $y \in [x-\delta_n,x+\delta_n]\cap K$.

Then we have $\lvert f_n(y) - f_n(x)\rvert \leqslant \varepsilon$ for all $y\in [x-\eta,x+\eta]\cap K$ and all $n$ if we choose

$$\eta = \min \{\delta\}\cup \{\delta_n : n < n_0\}.$$

Thus $\{ f_n\}$ is equicontinuous in $x$. Since $x$ was arbitrary, $\{f_n\}$ is equicontinuous on $K$. Since $K$ is compact, $\{f_n\}$ is uniformly equicontinuous on $K$, and therefore the topologies of pointwise convergence and uniform convergence on $K$ coincide on $\{f_n\}$.

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.