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

Is it possible to decompose a discrete-time martingale $(M_n)$ uniquely into two processes $$M_n=M_n^I+A_n$$ where $(M^I_n)$ is a martingale with independent increments and $(A_n)$ is a martingale? If no, under which condition(s) on $(M_n)$, do we have this kind of decomposition?

share|cite|improve this question
Yes. (Hint: Let $M_n^I$ be something very trivial; or, if you want something slightly less trivial, add and subtract.) – cardinal Oct 10 '12 at 0:19
For the first (very trivial) part, yes, $M_n^I = 0$ is valid. For the second (only slightly less trivial) part, add and subtract a martingale $(S_n)$ with independent increments that is independent of $(M_n)$. – cardinal Oct 10 '12 at 0:39
I edited my question. Actually what I mean is an unique decomposition similar to Doob decomposition. – Martungale Martun Oct 10 '12 at 0:44
Then, the answer is obviously "no", by considering examples in the same spirit as the ones I gave. – cardinal Oct 10 '12 at 0:46
thanks cardinal. I again edited the question. I would like to learn under which conditions it is possible to do. Any reference would also be helpful. – Martungale Martun Oct 10 '12 at 0:55

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.