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

Let $f$ be in $L^{1}(\mathbb{R})$, $\mathbb{R}$ the real numbers. Show that for every $\varepsilon > 0$ there exists $A \subseteq R$ , measurable, such that $m(A) < \infty$ , $f$ is bounded on $A$ and $ \int_{\mathbb{R}} |f| < \int_{A} |f| + \varepsilon$.

If we take $A$ as the support of the simple function which approximates $f$ in the $L^{1}$ norm then this has finite measure and it satisfies the other conditions. But I don't see why $f$ must be bounded on it. Any ideas?

Thank you.

share|cite|improve this question
@user100: take a sequence of functions converging pointwise to $f$ which are bounded (e.g., $f_n$ equal to $f$ if the values are between $-n$ and $n$, and either $-n$ or $n$ otherwise), then approximate $f$ with some bounded function and the bounded function with a simple function. I haven't worked out the details, so this may very well crash and burn... – Arturo Magidin Nov 6 '10 at 20:03

$|f|I_{|f|\leq n} $ increases to $|f|I_{|f| < \infty} = |f|$ (a.e.) as $f$ is in $L^{1}(\mathcal{R})$, the result follows from monotone convergence theorem.

Update : This answer is incomplete, please see Jonas's answer.

share|cite|improve this answer
Sorry for the overlap. I was still getting the TeX to work when your answer was posted. Your answer would have to be modified to get $m(A)\lt\infty$. – Jonas Meyer Nov 6 '10 at 20:38

You can do this as a "$2\epsilon$-proof" (or $\epsilon/2$ if you prefer). First, since $(\int_{-n}^n|f|)_{n=1}^\infty$ converges to $\int_{\mathbb{R}}|f|$, there is an $n$ such that $\int_{\mathbb{R}}|f|\lt\int_{-n}^n|f|+\epsilon/2$. Then, since $(\int_{-n}^n|f|\cdot\chi_{\{|f|\leq m\}})_{m=1}^\infty$ converges to $\int_{-n}^n|f|$, there is an $m$ such that $\int_{-n}^n|f|\lt\int_{-n}^n|f|\cdot\chi_{\{|f|\leq m\}}+\epsilon/2$. Take $A=[-n,n]\cap\{x:|f(x)|\leq m\}$.

Rather than using simple functions to show this, I would use this as a first step to showing that $f$ can be approximated by simple functions, because now $f$ can be uniformly approximated by simple functions on $A$.

share|cite|improve this answer
(The 2 steps aren't really necessary, and you could just use the fact that $|f|\cdot \chi_{\{x\in[-n,n]:|f(x)|\leq n\}}$ converges to $|f|$ pointwise a.e.) – Jonas Meyer Nov 6 '10 at 20:23
Nice, thank you Jonas. – student Nov 7 '10 at 1:05
@user100: If you like it you should approve the answer :) – AD. Nov 7 '10 at 13:05
@user100: You're welcome. Of course this is just a sketch, and justification is needed for the 2 claims of convergence. For example, monotone or dominated convergence can be used. – Jonas Meyer Nov 7 '10 at 18:04

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.