Welcome to MathOverflow

Question

0

4

Hi Everyone

I have some difficulties deriving the Stochastic Differential Equation for the following problem, any help or reference would be appreciated.

We are given a Brownian Motion $B_t$ and we note $M_t=\sup_{s\le t}B_s$ . Moreover we have a smooth real valued function $F(t,x,y)$ (for example a $C^{1,2,1}$) over $\mathbb{R}^+\times \mathbb{R} \times \mathbb{R}^+$ and we are looking for the SDE followed by $F(t,M_t-B_t,M_t)$

I have a hard time trying to express $dF$.

Best regards

edited Apr 21 2010 at 14:04

Harald Hanche-Olsen
5,750●1●13●34

asked Apr 21 2010 at 11:44

The Bridge
634●4●8

	After two seconds of reflection the answer is simple. It suffices to view $F$ as a function of $B_t$ and of the two increasing functions $M_t$ and $t$ and to apply Itô's lemma (Revuz and Yor bok's version for example). Sorry for this post, I first thought some Itô-Tanaka-Meyer formula were involved because of the fact that the processes $(L_t,\|W_t\|)$ (L is the local time of W at 0) and $(M_t-B_t,M_t)$ have the same law. Regards – The Bridge Apr 22 2010 at 8:43
	The Bridge: The MO bot bumped your question since in the system it remains "unanswered". Since zhoraster did provide an answer, it would be polite to accept it and award him or her the points. Plus, this will be useful for future generations of students who may ask the same question as you did. – Tom LaGatta Sep 8 2010 at 15:31

zhoraster · Accepted Answer · 2010-08-04 10:38:15Z UTC

1

Let $F(t,m-b,m)=G(t,m,b)$ (this way it's easier to write). As long as $M$ has bounded variation, we can happily write (I skip arguments, which are $t,M_t,B_t$) $$ dG(t,M_t,B_t) = G'_t dt + G'_m dM_t + G'_b dB_t + \frac12 G''_{bb} dt. $$

answered Aug 4 2010 at 10:38

zhoraster
558●4●11

Sorry but I already answered the question myself in the comment. – The Bridge Aug 5 2010 at 7:50

1

Yes, I didn't read it carefully. Still I don't understand why downvote my answer. – zhoraster Aug 17 2010 at 8:47

As Tom LaGatta suggested I did tagged the answer as "Accepted" – The Bridge Sep 8 2010 at 15:51

Looking for a version of Itô's Lemma

Remember to vote up questions/answers you find interesting or helpful (requires 15 reputation points)

1 Answer

You can accept an answer to one of your own questions by clicking the check mark next to it. This awards 15 reputation points to the person who answered and 2 reputation points to you.