Good day, I have the next theorem: Let {${E_i}$} a collection denumerable of measurable sets such that $E_1\supset E_2\supset... \supset E_n\supset... $ and $m(E_1)$ is finite, then $m(\bigcap_{i=1}^{\infty}E_i)=lim_{n\rightarrow \infty}m(E_n)$.

How is the theorem false if $m(E_1)=\infty$, with a counterexample? I think in intervals $I_k$ in some set A such that $I_k=(a,b)$, $a,b \in \mathbb{Q}$ . Can that perform?

  • $\begingroup$ every interval $I_k$ in your question @mathreda has a finite measure so you cannot create a counterexample with these $I_k$ $\endgroup$ – Marios Gretsas Aug 19 '17 at 22:12

The standard counterexample is to set $E_n=(n,\infty)$ for each $n$. Then $m(E_n)=\infty$ for all $n$, but $\bigcap_{n=1}^{\infty}E_n=\emptyset$.

| cite | improve this answer | |
  • 1
    $\begingroup$ Hell nooo! 1 second before i pressed <Post your answer> you posted...:P...+1 from me though.(after two hours because im blocked from voting) $\endgroup$ – Marios Gretsas Aug 19 '17 at 21:56
  • $\begingroup$ It happens to all of us... $\endgroup$ – carmichael561 Aug 19 '17 at 21:56
  • $\begingroup$ What about a counterexample where only the first $E_1$ is infinite? $\endgroup$ – mrp Aug 19 '17 at 21:58
  • $\begingroup$ @mrp: The result in the question is true provided that there is some $N$ such that $m(E_n)<\infty$ for $n\geq N$. So there is no such counterexample. $\endgroup$ – carmichael561 Aug 19 '17 at 21:59
  • $\begingroup$ @carmichael561 alright, that makes intuitive sense to me, although that's not what the OP stated. $\endgroup$ – mrp Aug 19 '17 at 22:03

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.