Moment estimates for solutions of SPDEs with Lévy colored noise
Raluca M. Balan, Juan J. Jiménez · 2025
Read Moment estimates for solutions of SPDEs with Lévy colored noise by Raluca M. Balan, Juan J. Jiménez (2025) 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-2504-21648.txt.
Introduction
Raluca Balan and Juan Jiménez bound the growth of noise-driven waves and heat on open ground, in a 2025 paper. Their equations are stochastic partial differential equations: part physics, part dice, with the dice colored by a spatial kernel and thrown by Lévy jumps, not smooth Gaussian noise. The key ask is whether the solution's moments, its spread at high powers, stay finite.
With a classic inequality for sums of independent kicks, they cap the p-th moment of the noise integral. That cap gives clean terms under which the linear heat and wave equations hold finite moments, for three kernels. The theory then stretches to nonlinear equations with tame coefficients. Growth proves at most exponential, at a rate they name, and some cases run weakly intermittent, bursting in rare spikes. For one model they expand the solution outright and write the second growth rate in full.
Catalog record
| CREATOR | Raluca M. Balan, Juan J. Jiménez |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 142 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2504-21648
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Moment estimates for solutions of SPDEs with Lévy colored noise Raluca M. Balan, Juan J. Jiménez Abstract In this article, we continue the investigations initiated by the first author in [ 2 ] related to the study of stochastic partial differential equations (SPDEs) with Lévy colored noise on ℝ + × ℝ d \mathbb{R}_{+}\times\mathbb{R}^{d} . This noise is constructed from a Lévy white noise on ℝ + × ℝ d \mathbb{R}_{+}\times\mathbb{R}^{d} (which is in turn built from a Poisson random measure on ℝ + × ℝ d × ℝ 0 \mathbb{R}_{+}\times\mathbb{R}^{d}\times\mathbb{R}_{0} with intensity d t d x ν ( d z ) dtdx\nu(dz) ), using the convolution with a suitable spatial kernel κ \kappa . We assume that the Lévy measure ν \nu has finite variance. Therefore, the stochastic integral with respect to this noise is constructed similarly to the integral with respect to the spatially-homogeneous Gaussian case considered in [ 14 ] . Using Rosenthal’s inequality, we provide an upper bound for the p p -th moment of the stochastic integral with respect to the Lévy colored noise, which allows us to identify sufficient conditions for the solution of an SPDE driven by this noise to have higher order moments. We first analyze this question for the linear SPDE (in which the noise enters in an additive way), considering as examples the stochastic heat and wave equations in any dimension d d , for three examples of kernels κ \kappa : the heat kernel, the Riesz kernel, and the Bessel kernel. Then, we present a general theory for a non-linear SPDE with Lipschitz coefficients, and perform a detailed analysis in the case of the heat equation (in dimension d ≥ 1 d\geq 1 ), and wave equation (in dimension d ≤ 3 d\leq 3 ), for the same kernels κ \kappa . We show that the solution of each of these equations has a finite upper Lyapounov exponent of order p ≥ 2 p\geq 2 , and in some cases, is weakly intermittent (in the sense of [ 19 ] ). In the case of the parabolic/hyperbolic Anderson model with Lévy colored noise, we provide the Poisson chaos expansion of the solution and the explicit form of the second-order Lyapounov exponent. MSC 2020: Primary 60H15; Secondary 60G60, 60G51 Keywords: stochastic partial differential equations, Poisson random measure, Lévy white noise, random field, Lyapounov exponent, intermittency, Poisson chaos expansion 1 Introduction The study of stochastic partial differential equations (SPDEs) is an active area of research in probability theory, which has been growing tremendously in the last four decades, since the publication of Walsh’ lecture notes [ 41 ] . Many investigations focus on equations driven by Gaussian noise, which may be white or colored in space or time. A significant portion of the literature is dedicated to equations driven by a pure-jump space-time Lévy white noise , which does not have a Gaussian component. This noise is given by a collection \{\Lambda(A);A\in\mathcal{B}_{b}(\mathbb{R}_{+}\times\mathbb{R}^{d})\} of random variables indexed by the bounded Borel subsets of \mathbb{R}_{+}\times\mathbb{R}^{d} , and is defined by: \Lambda(A)=b|A|+\int_{A\times\{|z|\leq 1\}}z\widehat{N}(dt,dx,dz)+\int_{A\times\{|z|>1\}}zN(dt,dx,dz), where b\in\mathbb{R} , N is a Poisson random measure (PRM) on \mathbb{R}_{+}\times\mathbb{R}^{d}\times\mathbb{R}_{0} of intensity \mu(dt,dx,dz)=dtdx\nu(dz) , \widehat{N}(F)=N(F)-\mu(F) is the compensated version of N , for Borel sets F in \mathbb{R}_{+}\times\mathbb{R}^{d}\times\mathbb{R}_{0} , and \nu is a Lévy measure on \mathbb{R} i.e. \nu(\{0\})=0 and \quad\int_{\mathbb{R}_{0}}(|z|^{2}\wedge 1)\nu(dz)<\infty. (1) In this theory, \mathbb{R}_{0}:=\mathbb{R}\verb 2\2\{0\} is equipped with the distance d(x,y)=|\frac{1}{x}-\frac{1}{y}| , so that “bounded sets” in \mathbb{R}_{0} are in fact those bounded away from 0 . One motivation for this comes from the symmetric \alpha -stable Lévy motion \{Z(t)\}_{t\in[0,1]} , which has Lévy measure: \nu_{\alpha}(dz)=\frac{1}{2}\alpha|z|^{-\alpha-1}1_{\{|z|>0\}}dz\quad\mbox{with}\quad\alpha\in(0,2), and for which the underlying PRM has points of the form (T_{i},Z_{i}=\varepsilon_{i}\Gamma_{i}^{-1/\alpha}) where (T_{i})_{i\geq 1} are i.i.d. uniform on [0,1] , (\varepsilon_{i})_{i\geq 1} are Rademacher variables, and (\Gamma_{i})_{i\geq 1} are the points of a Poisson process on [0,\infty) of intensity 1. Since |Z_{i}|\to 0 as i\to\infty , and a PRM has a finite number of points in any compact set, we must exclude 0 from the “jumps space”. The moments of \Lambda(A) are closely related to integrability properties of \nu : for any p>0 , \mathbb{E}|\Lambda(A)|^{p}<\infty if and only if \int_{|z|>1}|z|^{p}\nu(dz)<\infty . Therefore, if we assume that m_{2}:=\int_{\mathbb{R}_{0}}|z|^{2}\nu(dz)<\infty, (2) then \Lambda(A) has finite variance. If in addition, we choose b=-\int_{|z|>1}z\nu(dz) , we obtain a centered process: L(A)=\int_{A\times\mathbb{R}_{0}}z\widehat{N}(dt,dx,dz),\quad A\in\mathcal{B}_{b}(\mathbb{R}_{+}\times\mathbb{R}^{d}). (3) Since \mathbb{E}\big[L(A)L(B)\big]=m_{2}|A\cap B|, the theory of stochastic integration with respect to L is the same as for the space-time Gaussian white noise W . Moreover, SPDEs with noise L share many similarities with the corresponding equations driven by W , with some notable exceptions, such as: the existence of higher-order moments, and the regularity of the sample paths. Throughout this work, we will assume that ( 2 ) holds, and we let L be defined by ( 3 ). We start by recalling the major contributions for SPDEs with general Lévy white noise, for which the techniques are very different compared to the Gaussian case. Then we return to the finite variance case, which is the topic of the present article. Note that for our literature review, we mention only the references based on Walsh’s random field approach. For a different approach to SPDEs with Lévy noise, we refer the reader to the monograph [ 36 ] by Peszat and Zabczyk, and the references therein. The stochastic heat equation with general Lévy noise: \frac{\partial u}{\partial t}(t,x)=\frac{1}{2}\Delta u(t,x)+\sigma\big(u(t,x)\big)\dot{\Lambda}(t,x),\quad t>0,x\in\mathbb{R}^{d} (4) was studied for the first time in [ 38 ] , where it was shown that if \sigma is Lipschitz, ( 4 ) has a unique solution, and this solution has a finite p -th moment, provided that p<1+\frac{2}{d} and m_{p}:=\int_{\mathbb{R}_{0}}|z|^{p}\nu(dz)<\infty. (5) Condition ( 5 ) is not satisfied by the \alpha -stable Lévy noise, with \alpha\in(0,2) . In [ 33 ] , Mueller proved the existence of a solution of ( 4 ) driven by an \alpha -stable Lévy noise with \alpha\in(0,1) , and a non-Lipschitz function \sigma(u)=u^{\gamma} with \gamma>0 . The same problem for the case \alpha\in(1,2) was treated by Mytnik in [ 34 ] . The compact support property for solutions to stochastic heat equations with \alpha -stable Lévy noise has been recently established in [ 23 ] . The problem of existence of a solution of a general SPDE with \alpha -stable Lévy noise and Lipschitz function \sigma remained open until 2014. In [ 1 ] , the first author of this paper established the existence of a solution of a large class of SPDEs on bounded domains, driven by the \alpha -stable Lévy noise. This was achieved by solving first the equation with truncated noise, obtained by removing the “large jumps” that exceed a flat level N . In 2017, Chong [ 11 ] , introduced a space-dependent truncation technique, and improved significantly the results of [ 1 , 38 ] by showing the existence of a solution of ( 4 ). His condition: \int_{\{|z|\leq 1\}}|z|^{p}\nu(dz)<\infty\quad\mbox{and}\quad\int_{\{|z|>1\}}|z|^{q}\nu(dz)<\infty\quad\mbox{for some $0<q\leq p$}, (6) uses different exponents p and q for the “small jumps” region \{|z|\leq 1\} versus the “large jumps” region \{|z|>1\} , and is satisfied by the \alpha -stable Lévy noise. Subsequently, the authors of [ 12 ] proved the regularity of the