8
$\begingroup$

Let $X$ be a continuous local martingale, and $\langle X \rangle$ be its quadratic variation process. The "standard" proof of Burkholder-Davis-Gundy inequalities found in books yields $(\mathsf{E} |X|^{p})^{1/p} \le O(p) \cdot (\mathsf{E} \langle X \rangle ^{p/2})^{1/p}$ for large $p$.

Can the growth rate be improved to, say, $O(p^{1/2})$? For example, if $\langle X \rangle$ is bounded, this estimate gives exponential tails for $|X|$, which is clearly suboptimal, since they should be Gaussian.

$\endgroup$
3
  • $\begingroup$ What is $\langle X \rangle$? $\endgroup$ May 31, 2012 at 20:49
  • $\begingroup$ Quadratic variation. Updated the post to clarify this. $\endgroup$ May 31, 2012 at 20:52
  • 5
    $\begingroup$ The best constants are known, and you can't do better than p-1 for p > 2. This was proven by Davis I think, but I'm not sure if that applies specifically to continuous martingales. $\endgroup$ May 31, 2012 at 21:56

3 Answers 3

7
$\begingroup$

I know a version which exactly gives the constant $O(p^{1/2})$ for $p\ge 2$. It is contained in a lecture note by D. Khoshnevisan on SPDE.

$\endgroup$
2
  • $\begingroup$ Do you mean Theorem 5.27 in math.utah.edu/~davar/ps-pdf-files/SPDE.pdf? On the bottom of p.17 he says, in his notation, something equivalent to $\left(\phi\left(t\right)\right)^{1/p}\le a_{p}^{1/2}\left(\mathsf{E}\left\langle N\right\rangle _{t}^{p/2}\right)^{1/p}$, where $\phi\left(t\right)=\mathsf{E}\sup_{\left[0,t\right]}\left|N\right|^{p}$ and $a_{p}=\frac{p\left(p-1\right)}{2}\left(\frac{p}{p-1}\right)^{p}$. This seems to give the $O(p)$ bound that I was talking about, not $O(p^{1/2})$. $\endgroup$ Jan 23, 2014 at 23:19
  • 3
    $\begingroup$ Please see p. 196 of the file: stt.msu.edu/CBMS2013/D_Khoshnevisan_Lecture.pdf $\endgroup$
    – epsilon
    Jan 24, 2014 at 14:40
1
$\begingroup$

You are correct that for bounded $<X>_T$ the tails of $X_T$ should be Subgaussian. However, the Burkholder-Davis-Gundy inequality gives an upper bound for the $L^p$-norm of the running supremum $X_T^* = \sup_{t \le T} |X_T|$, of $X$ not just for $X_T$ itself.

I do not see a reason why $X_T^*$ should have Subgaussian tails, even if $<X>_T$ is bounded. In fact it cannot always have Subgaussian tails, otherwise the known optimal constant $p-1$ for $p \ge 2$ (see George Lowthers remark) would not be optimal.

$\endgroup$
4
  • 1
    $\begingroup$ The question was about continuous martingales. A continuous martingale is a time-changed Brownian motion, and the maximum of a Brownian motion over a bounded interval is sub-Gaussian. $\endgroup$ Nov 13, 2015 at 14:53
  • $\begingroup$ The time-change may be random and unbounded, thus your argument regarding Brownian Motion does not transfer to general continuous martingales. $\endgroup$
    – MKR
    Nov 17, 2015 at 10:05
  • 1
    $\begingroup$ The time change is the quadratic variation. I thought you were talking about the case when it's bounded. $\endgroup$ Nov 17, 2015 at 14:02
  • $\begingroup$ Yes, sorry, in the context of my answer your comment makes perfect sense, of course. $\endgroup$
    – MKR
    Nov 17, 2015 at 15:29
1
$\begingroup$

This comment was too long.

For martingales of the form $X_{t} = \int_{0}^{t}f(s,\omega)dB_{s}$ where $B_{t}$ is the standard Brownian motion, $f(s, \omega)$ is non-anticipating function such that $\int_{0}^{\infty}f^{2}(s, \omega)ds <\infty$ a.s. (I guess they are called $L^{2}$ integrable continuous martingales) one can obtain the bound of order $\sqrt{p}$ for large $p$. This is due to Burgess Davis, (see Section 3) of the reference. The result essentially follows from the Brownian case with arbitrary $L^p$ integrable stopping times.

However, for discrete time martingales, this is not true, and the best possible bound is of order $p$ for large $p$, for example, see paper of G. Wang, Remark 2. On the other hand, there are very special ``conditionally symmetric'' discrete martingales for which one can obtain the bound of order $\sqrt{p}$. And it looks like continuous $L^{2}$ integrable martingales in this sense are similar to conditionally symmetric ones.

Wang, Gang, Sharp inequalities for the conditional square function of a martingale, Ann. Probab. 19, No.4, 1679-1688 (1991). ZBL0744.60046.

Davis, Burgess, On the $L^p$ norms of stochastic integrals and other martingales, Duke Math. J. 43, 697-704 (1976). ZBL0349.60061.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.