Take the 2-minute tour ×
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.

Suppose $B$ is the standard Brownian motion, how to calculate $$\mathbb{P}\left((\max_{0\le s\le t}B(s))-B(t)<a\right)$$ I tried $B(t)-B(s)=B(t-s)$, $$P(B(t-s)>-a)=\int_{-a}^{\infty}\frac{1}{\sqrt{2\pi(t-s)}}e^{-\frac{y^2}{2(t-s)}}dy$$ And then I integrate it via $s$, that is $$\mathbb{P}\left((\max_{0\le s\le t}B(s))-B(t)<a\right)=\int_{0}^{t}\int_{-a}^{\infty}\frac{1}{\sqrt{2\pi(t-s)}}e^{-\frac{y^2}{2(t-s)}}dy ds$$ Am I right in this process? I am novice to stochastic process, and I don't know if these manipulations are legal. Also, I don't know how to integrate this stuff. Thanks for your attention!

share|improve this question

1 Answer 1

up vote 4 down vote accepted

You need a couple of tricks for this one.

First use that fact that if $s\mapsto B(s)$ is a Brownian motion then $s\mapsto B(t-s)-B(t)$ is also a Brownian motion. Hence

$$\mathbb{P}\left((\max_{0\le s\le t}B(s))-B(t)<a\right)= \mathbb{P}\left(\max_{0\le s\le t}B(s)<a\right)$$

Now we can use the reflection principle.

Let $T_a$ be the first hitting time of $a$ and set $$B_a(s) = \begin{cases}B(s) &\mathop{ if } s\leq T_a \\ a-B(s) &\mathop{ if } s\geq T_a \end{cases}$$

Now $B_a(s)$ is also a Brownian motion and $\max_{0\le s\le t}B(s)<a$ if and only if both $B(t) < a$ and $B_a(t) < a$.

As the events $B(t) \geq a$ and $B_a(t) \geq a$ are mutually exclusive (up to an event of probabililty $0$) and have identical probabilities we have that

$$\mathbb{P}\left((\max_{0\le s\le t}B(s))-B(t)<a\right)= 1 - 2\mathbb P\left( B(t)\geq a\right).$$

Which is easy as $B(t)$ is normally distributed.

share|improve this answer

Your Answer

 
discard

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.