1
$\begingroup$

I have a question about the following post

$P[X_n\:\mathrm{converges}] = 0$ for iid non-constant RVs

As far as I can understand, we have by Borell-Cantelli Lemma that $$P(X_n \leq x \text{ infinitely often }) = 1= P(X_n \geq y \text{ infinitely often }) $$ from which it is clear that $$P(\liminf X_n \leq x) = 1 = P(\limsup X_n \geq y) $$

But, then how does one get

$$P(\liminf X_n \leq x \text{ and } \limsup X_n \geq y) = 1\ ? $$

$\endgroup$
  • $\begingroup$ That is the one of the most vague titles possible. $\endgroup$ – Akiva Weinberger Oct 2 '15 at 1:00
  • $\begingroup$ @AkivaWeinbergercolumbus I will try to come up with a better title $\endgroup$ – user74261 Oct 2 '15 at 1:03
  • $\begingroup$ @pm021 To-do list: 1. Switch to nonempty titles. 2. When asking about details of a proof on a given page, mention the new question on the old page, if only by correction. $\endgroup$ – Did Oct 2 '15 at 6:35
2
$\begingroup$

If $P(A)=P(B)=1$, then $P(A\cap B)=1$. To check this notice that $P((A\cap B)^c)=P(A^c\cup B^c)\le P(A^c)+P(B^c)$.

$\endgroup$

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.