1
$\begingroup$

Why is $f_{M_t,W_t}(m,w) = \frac{2 ( 2 m - w)}{t\sqrt{2 \pi t}} e^{-\frac{(2m-w)^2}{2t}}, m \ge 0, w \leq m$ ? I now know what running maximum is, but unsure why joint distribution goes as above formula.

$\endgroup$
0
$\begingroup$

This is a consequence of the reflection principle: $$ \mathbb{P}\left(B_t \leqslant x, M_t \geqslant y\right) = \mathbb{P}\left(B_t \leqslant x, T_y \leqslant t \right) \stackrel{\text{refl.pr.}}{=} \mathbb{P}\left(\hat{B}_t \geqslant 2y-x, T_y \leqslant t \right) = \mathbb{P}\left(B_t \geqslant 2 y -x, T_y \leqslant y\right) $$ where the last equality holds because the hitting time $T_y$ is the same for $B_t$ and the reflected Brownian motion $\hat{B}_t$. Since $y>x$ and $\{B_t \geqslant 2y-x\} \subset \{T_t \leqslant t\}$ we have $$ \mathbb{P}\left(B_t \leqslant x, M_t \geqslant y\right) = \mathbb{P}\left(B_t \geqslant 2 y -x\right) $$ It now remains to differentiate to recover the probability density function.

Added: See F.C. Klebaner, "Introduction to Stochastic Calculus with Applications", 3rd ed., ch 3.6 for more details.

$\endgroup$

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.