Let W(t) be a standard Brownian Motion starting from zero and M(t) be its running maximum, M(t)=max B(s), 0≤s≤t
For fixed time t compute the density of random variable M(t)-B(t)
ok, I will give a breif solution,
1. For any two random variables X and Y (in our case, M(t) and B(t)), given their joint density f(X,Y), what's the density of their difference Z=X-Y?
The distribution function of Z, F(z)=P(Z<=z)=P(X-Y<=z)=∫∫_D f(x,y) dxdy, where D={(x,y): x-y<=z and x>=y, x>=0}, (note x>=y is for our special case M(t)>=B(t), and since B starts from 0, M(t)>=0)
F(z)=∫_0^inf[ ∫_x-z^x f(x,y) dy]dx
f(z)=F'(z)=∫_0^inf f(x,x-z) dx
2. Check your text book, the joint density:
\[f(x,y)=\frac{2(2x-y)}{t\sqrt{2\pi t}}e^{-\frac{(2x-y)^2}{2t}}\]