2
$\begingroup$

We define

$p_T(f):= \mathbb{P}(\forall_{s\leq T}B_s \geq f(s)-1),$

where $B$ is a Brownian motion such that $B_0 =0$ and $f:\mathbb{R}_+ \mapsto \mathbb{R}$ is some (continuous) function. I am interested in understanding the random variable:

$X_T=p_T(\tilde{B}),$

where again $\tilde{B}$ is a Brownian motion (I use $\tilde{B}$ just to avoid any confusion). What is easy

$\mathbb{E}X = \mathbb{P}(\forall_{s\leq T}B_s \geq \tilde{B}_s-1) \sim T^{-1/2}.$

I would be interested in knowing if e.g.

$\mathbb{P}(X_T \geq T^{-A})\rightarrow 1,$

for $A$ large enough.

$\endgroup$
2
  • 1
    $\begingroup$ Have you tried expressing $X$ as the expected crossing time of a Brownian motion after subtracting the two motions and rescaling? $\endgroup$ Dec 7, 2012 at 16:44
  • $\begingroup$ @Steve - I am not quite sure what you mean. But I do not see any "easy, standard" way to use Brownian scaling. $\endgroup$ Dec 8, 2012 at 12:16

0

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.