Optimal stopping and best constants for Doob-like inequalities. I: The case p=1
Optimal stopping and best constants for Doob-like inequalities. I: The case \(p=1\)
Let \(X\) be a non-negative, uniformly integrable submartingale \(X\in D(0,\infty)\). Define \(X^*_ t=_{s\leq t}X_ s\). The best constant \(c_ q\) appearing in the inequalities of the form \(EX^*_ \infty\leq c_ q| X_ \infty|_ q\), \(q>1\), is established. It is shown that the best possible constant \(c_ q=(\Gamma(1+\tilde q))^{1/\tilde q}\), where \(\tilde q=q/(q-1)\). The method of proof is via explicitly solving the problem of optimal stopping the process of the form \(M^*_ t-f(| M_ t|)\), where \(f\) is \({\mathfrak R}\cup\{+\infty\}\)-valued function. Inequalities bounding \(EX^*_ \infty\) by suitable multiples of the Luxemburg and Orlicz norms of \(X_ \infty\) with respect to a convex Young function are also obtained. The result is applied to a version of John-Nirenberg inequality for martingales of bounded mean oscillation. By using a different method the same result concerning the best constant \(c_ q\) was obtained by \textit{D. Gilat} [Isr. J. Math. 63, No. 3, 270-280 (1988; Zbl 0661.60060)].
- Doob's inequalities revisited: A maximal H^ 1-embedding
- The root solution to the multi-marginal embedding problem: an optimal stopping and time-reversal approach
- Three-dimensional Brownian motion and the golden ratio rule
- Perpetual American options in diffusion-type models with running maxima and drawdowns
- A generalization of the Burkholder–Davis–Gundy inequalities
- Stopping at the maximum of geometric Brownian motion when signals are received
- The Best Bound in the L logL Inequality of Hardy and Littlewood and its Martingale Counterpart
- scientific article; zbMATH DE number 4098413 (Why is no real title available?)
- Quickest detection of a hidden target and extremal surfaces
- Robust bounds for forward start options
- Sharp maximal inequalities for stochastic processes
- Watermark options
- scientific article; zbMATH DE number 4184518 (Why is no real title available?)
- A note on the stochastic version of the Gronwall lemma
- The support and resistance line method: an analysis via optimal stopping
- On the drawdowns and drawups in diffusion-type models with running maxima and minima
- An explicit solution for an optimal stopping/optimal control problem which models an asset sale
- Pathwise inequalities for local time: Applications to Skorokhod embeddings and optimal stopping
- \(\pi \) options
This page was built for publication: Optimal stopping and best constants for Doob-like inequalities. I: The case \(p=1\)
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1180587)
