Space-Time Transport of Brownian Exit Laws
Maher Boudabra · 2026
Read Space-Time Transport of Brownian Exit Laws by Maher Boudabra (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2605-25742.txt.
Introduction
Boudabra gives a systematic study of p-Brownian convergence, the notion he introduced to study planar Skorokhod embedding stability. The first part shows shape aspects of it. The second part turns the notion into a metric between domains, placed within optimal transport theory. Results include long-run trends for scaled domains, with computed cases. The paper thus turns it from a stability tool into a full metric scheme for domains.
Catalog record
| CREATOR | Maher Boudabra |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 81905 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.25742
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Space-Time Transport of Brownian Exit Laws Maher Boudabra Abstract The present paper is devoted to a systematic study of the p p -Brownian convergence introduced in [ 5 ] (in press) to study the stability of the planar Skorokhod embedding problem [ 14 , 6 ] . The first part is an illustration of some geometric aspects of the p p -Brownian convergence. The second part turns this notion into a metric between domains. More precisely, we place it within the framework of optimal transport theory. Several results are obtained, namely asymptotic behavior in case of homothetic domains. Numerical illustrations are provided as well. 1 Preliminaries In a recent work [ 5 ] , motivated by the study of numerical approximation of the solutions of the planar Skorokhod embedding problem [ 14 , 8 , 6 ] , the authors introduced a new mode of convergence for planar domains, which they called p -Brownian convergence with p>0 . This mode of convergence captures the geometry of domains through the time-space exit pair of planar Brownian motion. We recall the definition in a form that emphasizes that only a coupling of the exit-pair laws is required. Definition 1 . Let U be planar domain and let (U_{n})_{n} be a sequence of planar domains, all containing a common point, say the origin. For each n , let (Z^{[n]}_{t})_{t\geq 0} be a planar Brownian motion started at 0 and denote by \tau_{n} its exit time from U_{n} . Similarly, let (Z_{t})_{t\geq 0} be a planar Brownian motion started at 0 and denote by \tau its exit time from U . We say that (U_{n})_{n} converges to a domain U in the p -Brownian sense if there exist two random vectors (\xi_{n},\theta_{n}) and (\xi,\theta) on some common probability space, such that (\xi_{n},\theta_{n})\sim(Z^{[n]}_{\tau_{n}},\tau_{n}),\,\,(\xi,\theta)\sim(Z_{\tau},\tau) and \begin{aligned} \mathbf{E}\left(\big|\xi_{n}-\xi\big|^{p}\right)+\mathbf{E}\left(\big|\theta_{n}-\theta\big|^{\frac{p}{2}}\right)&\underset{n\to+\infty}{\longrightarrow}0\end{aligned}. This formulation should be understood as a coupling of the exit-pair laws only. It does not require the Brownian motions in U_{n} and U to be realized as the same planar Brownian path. Same-path couplings will appear later as a useful special case, but they are not part of the definition. Note that p -Brownian convergence implies q -Brownian convergence for 0<q<p . The joint realization of the two random vectors (\xi_{n},\theta_{n})\qquad\text{and}\qquad(\xi,\theta) is called a coupling of the exit-pair laws of (Z^{[n]}_{\tau_{n}},\tau_{n})\qquad\text{and}\qquad(Z_{\tau},\tau). This will be elucidated in section 3 . The pair (Z_{\tau_{U}},\tau_{U}) consists of the Brownian exit location and the Brownian exit time. The first component records where the boundary is seen from the base point, while the second records the time scale of escape. Taken together, these two variables provide a natural probabilistic signature of U . In this sense, Brownian convergence measures the effective geometry of a domain as seen by a diffusing particle starting from the origin. The law of Z_{\tau} is the harmonic measure \omega^{0}_{U} of the domain U . If the boundary of U is smooth, then in many cases, one can manage to compute the probability density \rho of Z_{\tau} . In this context, when U is simply connected then univalent functions are widely used to compute the p.d.f of Z_{\tau} . More precisely, if f is a univalent map from \mathbb{D} onto U , with f(0)=0 , that extends to be analytic across \partial\mathbb{D} then d\omega^{0}_{U}(z)=\frac{1}{2\pi}\bigl|(f^{-1})^{\prime}(z)\bigr|\,|dz|,\qquad z\in\partial U. (1.1) See [ 21 ] . On the other hand, the law of the exit time \tau is usually much harder to compute explicitly. To the best of our knowledge, there are only a few domains for which the density of \tau is explicit. The best-known example is the half-plane, where the problem reduces to the hitting time of a one-dimensional Brownian motion. Since the survival function (z,t)\mapsto\mathbf{P}_{z}(\tau_{U}>t) solves the heat equation with Dirichlet boundary condition, PDE techniques are often used to study the law of \tau_{U} . For example, one obtains infinite-series representations when U is the unit disc or an infinite strip. The literature also contains estimates for more challenging domains [ 26 , 1 ] . Thus in our framework, the joint law of the pair (Z_{\tau},\tau) is hard to compute explicitly. Moreover, Z_{\tau} and \tau are typically not independent. Definition 2 . Let p>0 . The Hardy space of index p , denoted by \mathbf{H}^{p}(\mathbb{D}) , is the space of all analytic maps f on the unit disc such that \sup_{0<r<1}\frac{1}{2\pi}\int^{2\pi}_{0}|f(re^{i\theta})|^{p}\,d\theta<+\infty. (1.2) The p -th root of 1.2 is denoted by \|f\|_{\mathbf{H}^{p}(\mathbb{D})} and it is called the p -th Hardy norm of f . When p<1 , the \|\cdot\|_{\mathbf{H}^{p}(\mathbb{D})} is not a “genuine” norm as the triangle inequality does not hold. A Hardy norm encodes how the magnitude of f varies on inner circles of the unit disc. We refer the reader to [ 27 , 13 ] for more details about Hardy spaces. As the unit circle is of finite measure, we have the inclusion \mathbf{H}^{p+q}(\mathbb{D})\subset\mathbf{H}^{p}(\mathbb{D}). This inclusion raises a natural inquiry about the supremum p for which a map f belongs to \mathbf{H}^{p}(\mathbb{D}) . Definition 3 . Let U\subsetneq\mathbb{C} be a simply connected domain, and let f:\mathbb{D}\to U be univalent onto U . The \emph{Hardynumber} of U is defined by h(U)=\sup\left\{p>0\mid\|f\|_{\mathbf{H}^{p}(\mathbb{D})}<\infty\right\}. The Hardy number h(U) is a well defined: that is, it does not depend on the choice of the conformal map f . It measures the largest exponent p for which a map f from \mathbb{D} onto U belongs to \mathbf{H}^{p}(\mathbb{D}) . Hardy number goes back to the work of Hansen [ 15 ] , who introduced this quantity in his study of Hardy classes and ranges of analytic functions. Since then, the Hardy number has been investigated from several geometric points of view, including descriptions in terms of harmonic measure and hyperbolic distance [ 16 ] . Its relevance to probability was clarified by the seminal work of D. Burkholder [ 9 ] , who showed that for a simply connected planar domain U the Hardy number governs the moments of the Brownian exit time. More precisely, if \tau_{U} denotes the first exit time of planar Brownian motion from U , then, among other results, Burkholder showed that if f:\mathbb{D}\to U is a univalent function onto U then \mathbf{E}(\tau^{\frac{p}{2}}_{U})<+\infty\Longleftrightarrow\|f\|_{\mathbf{H}^{p}(\mathbb{D})}<+\infty\Longleftrightarrow\mathbf{E}\left(\sup_{0\leq s\leq\tau_{U}}|Z_{s}|^{p}\right)<+\infty. (1.3) In particular, h(U)=2\sup\left\{q>0:\ \mathbf{E}(\tau^{\,q}_{U})<\infty\right\}. (1.4) As the right-hand side of 1.4 is meaningful for any planar domain, we adopt it as the definition of the Brownian Hardy number: h(U):=2\sup\left\{q>0:\ \mathbf{E}(\tau^{q}_{U})<\infty\right\}. This extension was introduced in [ 19 ] under the name Brownian Hardy number. The Hardy number encodes the heaviness of the tail of \tau_{U} . If U is bounded then obviously h(U)=+\infty . If the domain U is simply connected then its Hardy number can not be too small. More precisely Theorem 4 . [ 13 ] If U is a proper simply connected domain then h(U)\geq\frac{1}{2} . In terms of Brownian exit times, Theorem 4 says that for any simply connected domain U \mathbf{E}(\tau^{\,q}_{U})<+\infty whenever q<\frac{1}{4} . The information is handy to describe the long time decay of the tail of \tau_{U} . By the layer cake representation of the expectation, we have \int^{+\infty}_{1}\frac{1}{t^{\gamma}}\mathbf{P}(\tau_{U}>t)dt<+\infty for any \gamma>\frac{3}{4} . In particular, no proper simply connected domain U can satisfy \mathbf{P}(\tau_{U}>t)\gtrsim t^{-\alpha}\qquad(t\to\infty) for some \alpha<\frac{1}{4} . In fact, the decay rate of \frac{1}{\sqrt[4]{t}} in