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 have the following exercise I wish to solve:

Let $(X,S,\mu)$ be a measure space s.t $\mu(X)=1$.

Let $\mu^{*}$be defined on $X$ by:

$\forall E\subseteq X:\,\mu^{*}(E):=\inf\{\sum_{i=1}^{\infty}\mu(A_{i})\,\mid\, A_{i}\in S,E\subseteq\cup A_{i}\}$

Let $E\subseteq X$ s.t $\mu^{*}(E)=1$, prove that if $A,B\in S,A\cap E=B\cap E$ then $\mu(A)=\mu(B)$

What I tried is to write $$A=(A\cap E)\cup(A\cap E^{c}),B=(B\cap E)\cup(B\cap E^{c})$$ and then apply $\mu^{*}$on both of them.

Since $$A\cap E=B\cap E$$ I would have had $$\mu^{*}(A\cap E)=\mu^{*}(B\cap E)$$ and I hoped that $$\mu^{*}(E)=1\implies\mu^{*}(E^{c})=0$$ and from $\mu^{*}$being monotone that means that $$\mu^{*}(A\cap E^{c})=\mu^{*}(B\cap E^{c})=0$$

hence $\mu^{*}(A)=\mu^{*}(B)$ but $A,B\in S$ hence $$\mu^{*}(A)=\mu(A),\mu^{*}(B)=\mu(B)$$ and than I would be done.

But from this post is turns out that $\mu^{*}(E^{c})$ doesn't have to be $0$ and so my argument is not valid.

Can someone please suggets how to solve this problem ? what I tried was my only idea.

share|cite|improve this question
Hint: Without loss of generality, assume that $A \subset B$ (otherwise replace $(A,B)$ by $(A, A \cup B)$. We need to show that $B - A$ has measure 0, and we know that it lies in the complement of $E$. Try to show that if $\mu ( B - A) > 0$, then together with $\mu^*(E) = 1$, it would imply that $\mu(X) > 1$. – user27126 Dec 17 '12 at 11:15
up vote 2 down vote accepted

If $\mu(A)<\mu(B)$, then $\mu(B\backslash A)>0$. Hence $\mu\big(X\backslash (B\backslash A)\big)<1$. Now $E\subseteq (X\backslash (B\backslash A)\big)$ since $A$ and $B$ agree on $E$. Hence $\mu^*(E)<1$.

share|cite|improve this answer
Are you assuming $A\subseteq B$ ? I don't understand what to do if that is not the case – Belgi Dec 17 '12 at 17:52
@Belgi No. $B=(B\backslash A)\cup(B\cap A)$ and hence $\mu(B)=\mu(B\backslash A)+\mu(B\cap A)$. Hence $\mu(B)-\mu(B\cap A)=\mu(B\backslash A)$. Now, $\mu(B\cap A)\leq \mu(A)$ , so $\mu(B)-\mu(A)\leq\mu(B\backslash A)$. – Michael Greinecker Dec 17 '12 at 18:02
Can you please explain why $E\subseteq (X\backslash (B\backslash A)\big)$ ? – Belgi Dec 17 '12 at 18:55
$A \cap E = B \cap E$ implies that $(B \backslash A )\cap E = \emptyset$, which implies that. – user27126 Dec 17 '12 at 20:41
@Belgi Suppose $x\in E$, but $x\notin (X\backslash (B\backslash) A)$. Then $x\in B\backslash) A$. Hence $x\in B\cap E$ and $x\notin A\cap E$, contradicting the assumption that $B\cap E=A\cap E$. – Michael Greinecker Dec 18 '12 at 8:28

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.