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

There is a proposition that states the following: Assume $E$ has finite measure. Let {$f_n$} be a sequence of measurable functions that converges pointwise a.e. on $E$ to $f$ and $f$ is finite a.e. on $E$. Then {$f_n$} converges in measure to $f$ on $E$. How do you show that this fails if $E$ has infinite measure?

share|cite|improve this question
By finding a counterexample. – cardinal Mar 28 '11 at 22:59
@cardinal: Do you mean like {$f_n$} = $\chi_{[n,n+1]}$ – Libertron Mar 28 '11 at 23:02
@Sachin: Thank you for editing. I've deleted my no longer relevant comments. @Whomever: In case anyone is wondering about the close votes, they were from before the question was edited. – Jonas Meyer Mar 28 '11 at 23:03
@Sachin: That is a good one. – Jonas Meyer Mar 28 '11 at 23:03
No this sequence does not converge uniformly to zero. It only converges pointwise. In fact $\Vert f_{n} - 0\Vert_\infty = 1$ for all $n$, while $f_{n}(x) = 0$ for all $n \geq x$ (that's what Jonas had in mind in his answer below). As I explained in my answer to your previous question, uniform convergence implies convergence in measure. Cardinal explained why convergence in measure fails. – t.b. Mar 28 '11 at 23:15
up vote 6 down vote accepted

To see that the proposition is not true in general if $E$ does not have finite measure, you just need a counterexample. My suggestion is to start with $E=\mathbb{R}$, and come up with a sequence of measurable functions $(f_n)_{n\geq 1}$ converging pointwise to $0$, while the measure of the set where $f_n(x)\geq1$ is always infinite. For instance, if you have $f_n(x)=0$ when $x<n$, then the sequence will converge to $0$ at each point regardless of what $f_n(x)$ is when $x\geq n$. This gives a lot of freedom for strange counterexamples.

share|cite|improve this answer

We make the comment of Sachin clearer. Choose $E=\mathbb{R}$ and $\mu$ is the Lebesgue measure on $E$. Consider the sequence of measurable functions $\{f_n(x)\}_{n\in \mathbb{N}}$ given by $$ f_n(x)=\begin{cases} 1 & n\leq x\leq n+1, \\ 0& \text{ortherwise}. \end{cases} $$ for all $x\in E$ and $n\in \mathbb{N}$. Then:

  • For every $x\in E$ we have $x\notin [n,n+1]$ or $f_n(x)=0$ for sufficiently large $n$ and so $$|f_n(x)-0|=|0-0|=0$$ for sufficiently large $n$. This implies that $$\displaystyle\lim_{n\rightarrow\infty}f_n(x)= 0$$ for all $x\in E$. Therefore the sequence $\{f_n(x)\}_{n\in \mathbb{N}}$ is pointwise convergent to $f=0$.

  • We observe that $$ \left\{x\in E: |f_n(x)-0|\geq \frac{1}{2}\right\}=[n, n+1]. $$ and so $$ \mu\left\{x\in E: |f_n(x)-0|\geq \frac{1}{2}\right\}=\mu([n, n+1])=1. $$ Hence $\{f_n(x)\}_{n\in\mathbb{N}}$ is not convergent in measure to $f=0$.

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.