Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

Working through my stochastic calculus script, I encountered the following identity, for which no proof is given: $ \langle M, N^\tau \rangle = \langle M, N \rangle^\tau $, if $ M, N $ are continuous local martingales, null at 0.

I know that (by uniqueness of the bracket) it suffices to show the following:

If both $ M,N $ are continuous local martingales and $ \tau $ is a stopping time, then $ N^\tau ( M - M^\tau ) $ is again a continuous local martigale.

But why is this true?

I know how to proof this with the properties of the stochastic integral. But since the above identity is used in our script during the construction of the stochastic integral, I would like to prove it directly. Does anybody know how this can be done?

Trying to prove it, I started like this:

$ M,N $ continuous local martingales $ \ \Rightarrow \ $ there are localizing sequences $ \tau_n, \sigma_n $ such that $ M^{\tau_n} $ and $ N^{\sigma_n} $ are bounded martingales. By the stopping theorem, $ N^{\tau_n \wedge \sigma_n \wedge \tau} $ and $ M^{\tau_n \wedge \sigma_n \wedge \tau} $ are bounded martingales, too.

Now I would like to proof that $ N^{\tau_n \wedge \sigma_n \wedge \tau} (M^{\tau_n \wedge \sigma_n } - M^{\tau_n \wedge \sigma_n \wedge \tau}) $ is a martingale, since this would then imply that $ N^\tau ( M - M^\tau ) $ is a continuous local martigale (with localizing sequence $ \tau_n \wedge \sigma_n $).

Thanks a lot for your help! Regards, Si

share|cite|improve this question
The fact that $N$ is a local martingale is irrelevant. Any continuous adapted process will work. – George Lowther Apr 9 '12 at 15:21
up vote 1 down vote accepted

Let $X^{(n)}=N^{\kappa_n}\cdot(M^{\varrho_n}-M^{\kappa_n})$, where $\varrho_n=\tau_n\wedge\sigma_n$ and $\kappa_n=\varrho_n\wedge\tau$ hence $\kappa_n\leqslant\varrho_n$. Then, $\mathrm dX^{(n)}_t=[\kappa_n\leqslant t\leqslant\varrho_n]\cdot N_{\kappa_n}\cdot\mathrm dM_t=[\kappa_n\leqslant t\leqslant\varrho_n]\cdot N^{\kappa_n}_t\cdot\mathrm dM_t^{\tau_n}$ and $M^{\tau_n}$ is a martingale hence $X^{(n)}$ is a martingale.

share|cite|improve this answer
Nice! Thanks a lot! – Mad Si Apr 9 '12 at 19:37

Your Answer


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.