Bismut-Elworthy-Li Formulae for Forward-Backward SDEs with Jumps and Applications

Jiagang Ren 1 , Hua Zhang 2 · 2025

Read Bismut-Elworthy-Li Formulae for Forward-Backward SDEs with Jumps and Applications by Jiagang Ren 1 , Hua Zhang 2 (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-2512-23288.txt.

Introduction

Jiagang Ren and Hua Zhang prove derivative formulas of Bismut-Elworthy-Li type for a class of forward-backward stochastic differential equations that include jumps. Such formulas express how a solution changes when its starting point shifts, without computing the derivative of the solution itself. Their equations are driven partly by a Poisson random measure, which models sudden random events.

To handle the jumps, they use the lent particle method developed by Bouleau and Denis, extending earlier results that lacked this tool. As an application, they obtain existence and uniqueness for a family of nonlocal quasi-linear integral partial differential equations whose solutions are differentiable in the space variable. The results hold even when the starting data and coefficients of those equations are rough.

Catalog record

CREATORJiagang Ren 1 , Hua Zhang 2
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE138 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2512-23288

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Bismut-Elworthy-Li Formulae for Forward-Backward SDEs with Jumps and Applications Thanks: This work is supported by National Natural Science Foundation of China (Grant Nos. 12261038 and 12371152), and Natural Science Foundation of Jiangxi Province (Grant Nos. 20232BAB201004 and 20242BAB23003). Jiagang Ren 1 , Hua Zhang 2 Abstract . Under nondegeneracy assumptions on the diffusion coefficients, we establish the derivative formulae of Bismut-Elworthy-Li’s type for forward-backward stochastic differential equations with respect to Poisson random measure using the lent particle method created by Bouleau and Denis, which is not given before. Applying this formula, the existence and uniqueness of a solution of nonlocal quasi-linear integral partial differential equations, which are differentiable with respect to the space variable, are obtained, even if the initial datum and coefficients of this equation are not. 2020 Mathematics Subject Classification 60H10, 60H07 1. Introduction In the present paper, we are concerned with the problem of derivative formulae of Bismut-Elworthy-Li’s type for the following forward-backward stochastic differential equations (SDEs in short) with respect to Poisson random measure: \displaystyle\begin{cases}X(\tau,t,x)=x+\int_{t}^{\tau}b(s,X(s,t,x))ds\\ \quad\quad\quad\quad\quad+\int_{t}^{\tau}\int_{{\mathcal{O}}}\sigma(s,X(s-,t,x))u\widetilde{N}(ds,du),\tau\in[t,T],\\ X(\tau,t,x)=x,\tau\in[0,t],\end{cases} (1) where b:[0,T]\times{\mathbb{R}}^{d}\rightarrow{\mathbb{R}}^{d} , \sigma:[0,T]\times{\mathbb{R}}^{d}\rightarrow{\mathbb{R}}^{d}\times{\mathbb{R}}^{l} , {\mathcal{O}}=\{u\in{\mathbb{R}}^{l};|u|\leqslant 1\}\setminus\{0\} , \widetilde{N} is the compensated process of some Poisson point process N on ({\mathcal{O}},{\mathcal{B}}({\mathcal{O}})) with intensity \nu , and \displaystyle\begin{split}&Y(\tau,t,x)+\int_{\tau}^{T}\int_{{\mathcal{O}}}Z(s-,t,x,u)\widetilde{N}(ds,du)\\ =&-\int_{\tau}^{T}\psi(s,X(s,t,x),Y(s,t,x),\int_{{\mathcal{O}}}Z(s,t,x,u)l(u)\nu(du))ds+\phi(X(T,t,x)),\end{split} (2) where \tau\in[0,T] , \psi:[0,T]\times{\mathbb{R}}^{d}\times{\mathbb{R}}\times{\mathbb{R}}\rightarrow{\mathbb{R}} and \phi:{\mathbb{R}}^{d}\rightarrow{\mathbb{R}} are given functions, and l:{\mathcal{O}}\rightarrow{\mathbb{R}} is a Borel function with growth condition |l(u)|\leqslant C(1\wedge|u|) , u\in{\mathcal{O}} . A derivative formula for Wiener functionals was first established by Bismut in [ 1 ] using Malliavin calculus on Wiener space. This result was subsequently extended by Elworthy and Li in their foundational work [ 10 ] , where they systematically derived formulae for the derivatives of heat semigroups—this contribution, together with Bismut’s earlier result, laid the core theoretical basis for what is now known as the Bismut-Elworthy-Li formulae. Unlike the Wiener space, there are several versions of Malliavin calculus on Poisson space which are not equivalent. In the recent years, quite a few papers appeared concerning the Bismut-Elworthy-Li formulae for SDEs with jumps, using various approaches, e.g., SDEs driven by general Poisson jumps processes by using stochastic diffeomorphism flows and Girsanov’s transformation ( [ 19 ] ), SDEs driven by subordinated Brownian motion by using conditional Malliavin calculus ( [ 22 , 21 ] ), SDEs driven by \alpha -stable-like noise by using Bismut’s approach( [ 20 ] ). In a series of works, Bouleau and Denis (see [ 4 ] and references therein) presented systematically a new method called the lent particle one which may be considered as a differential calculus on Poisson space completely comparable to the Malliavin differential calculus on the Wiener space. The main feature of this method is its simplicity and convenience when applied to Poisson functionals, especially to solutions of stochastic differential equations driven by Lévy processes. Using this method, the Bismut-Elworthy-Li formulae for SDEs with jumps is also established in our recent paper [ 17 ] which generalizes and improves some previous works in this respect. Although some authors have studied the Bismut-Elworthy-Li formulae for SDEs with jumps, no one has yet discussed the same problem for forward-backward SDEs with jumps. For the cases of forward-backward SDEs without jumps, under nondegeneracy assumptions on the diffusion coefficient (that may be nonconstant), Fuhrman and Tessitore [ 11 ] have proved an analogue of the Bismut-Elworthy-Li formulae. After that, Masiero [ 16 ] has also obtained a Bismut-Elworthy-Li formula for forward-backward SDEs in a Markovian framework when the generator has quadratic growth with respect to Z . In the present paper, using the lent particle method, we establish a derivative formula of Bismut-Elworthy-Li’s type for forward-backward SDEs with respect to Poisson random measure under the nondegeneracy assumption on the diffusion coefficients, which has not been given so far to our knowledge, and we apply this formula to the study of nonlocal quasi-linear integral partial differential equations (PDEs in short). The organization of this paper is as follows. In Section 2, we recall the lent particle method which is the basic tool in our article. In Section 3, some notations and preliminary results on forward-backward SDEs are introduced. In Section 4, we obtain a derivative formula of Bismut-Elworthy-Li’s type for forward-backward SDEs with jumps. Building upon this formula, in Section 5, we establish the existence and uniqueness of solutions to nonlocal quasi-linear integral-PDEs that are differentiable with respect to the space variable, regardless of the potential non-differentiability of the initial conditions or coefficients of the equation. In the present paper, let C_{p} be a positive constant only depending on some parameter p , whose value may change from line to line. When we don’t want to emphasize this dependence we just use C instead. 2. Lent Particle Method 2.1. Set-up Let us first specify the general set-up in which we will work. We follow [ 4 ] to which we refer for more details. 2.1.1. Dirichlet Structure on the Bottom Space We start from a bottom space (\Xi,{\mathcal{G}},\nu) , where \Xi is a separable Hausdorff space, {\mathcal{G}} its Borel \sigma -algebra and \nu a \sigma -finite and diffuse measure on (\Xi,{\mathcal{G}}) . Let (\mathbf{d},e) be a local symmetric Dirichlet form on L^{2}(\nu) which admits a carré du champ operator \gamma . That is to say, \gamma is the unique positive, symmetric and continuous bilinear form from \mathbf{d}\times\mathbf{d} to L^{1}(\nu) such that \displaystyle e(f,g)=\frac{1}{2}\int_{\Xi}\gamma[f,g]d\nu,\quad\forall f,g\in\mathbf{d}. Moreover, we suppose that there exist \{k_{n},n\in{\mathbb{N}}\}\subset\mathbf{d} and A_{n}\uparrow{\mathcal{X}} such that k_{n}1_{A_{n}}\uparrow 1 and \gamma[k_{n}]1_{A_{n}}=0 . The structure (\Xi,{\mathcal{G}},\nu,\mathbf{d},\gamma) is called the bottom structure. Since \mathbf{d} is separable, the bottom Dirichlet structure admits a gradient operator, i.e., there exists a separable Hilbert space H and a linear map D from \mathbf{d} into L^{2}(\nu;H) such that \displaystyle\gamma[u]=\|Du\|_{H}^{2},\quad\forall u\in\mathbf{d}. Let (R,{\mathcal{R}},\rho) be another probability space such that the vector space L^{2}(R,{\mathcal{R}},\rho) is infinite dimensional. Take H=L_{0}^{2}(R,{\mathcal{R}},\rho)=\{g\in L^{2}(R,{\mathcal{R}},\rho);\int_{R}g(r)\rho(dr)=0\} . The corresponding gradient will be denoted by \flat , and we assume without any loss of generality that constants belong to \mathbf{d}_{loc} (see [ 5 , Chapter I, Definition 7.1.3] ) and so that 1^{\flat}=0 . 2.1.2. Space-time Setting and Dirichlet Structure on the Upper Space From now on we set X=[0,T]\times\Xi , {\mathcal{X}}={\mathcal{B}}([0,T])\times{\mathcal{G}} and \mu=dt\times\nu . Define the Dirichlet structure on (X,{\mathcal{X}},\mu) to be the product of the trivial one on (L^{2}([0,T],dt),0) and (\mathbf{d},e) and we keep the same notations \mathbf{d} , e , \flat , \gamma and \mathbf{a} , etc, for operators corresponding to this new Dirichlet form but note that they act only on the second variable. It is known that X is totally ordered (see [ 8 , Section 1, Theorem 11] ) and we denote by \prec such a total order relation. Set \displaystyle\Omega:=\{\omega=\sum_{i=1}^{\infty}\varepsilon_{y_{i}};\quad y_{i}\in X,\quad\forall i,\quad\text{and}\quad y_{1}\prec y_{2}\prec\cdots\prec y_{n}\prec\cdots\}. Let N be the Poisson random measure with intensity \mu defined on (\Omega,\mathcal{F},{\mathbb{P}}) where N(\omega)=\omega , \mathcal{F} is the \sigma -algebra generated by N and {\mathbb{P}} the law of N . Define \{\mathcal{F}_{t},t\in[0,T]\} to be the {\mathbb{P}} -complete right continuous filtration generated by \{N([0,t]\times G),t\in[0,T],G\in{\mathcal{G}}\} . We now introduce the creation and annihilation operator \varepsilon^{+} and \varepsilon^{-} : \displaystyle\forall(t,y)\in{\mathbb{R}}^{+}\times\Xi,\quad\forall\omega\in\Omega, \displaystyle\varepsilon_{(t,y)}^{+}(\omega)=\omega 1_{\{(t,y)\in supp\omega\}}+(\omega+\varepsilon_{(t,y)})1_{\{(t,y)\notin supp\omega\}}, \displaystyle\forall(t,y)\in{\mathbb{R}}^{+}\times\Xi,\quad\forall\omega\in\Omega, \displaystyle\varepsilon_{(t,y)}^{-}(\omega)=\omega 1_{\{(t,y)\notin supp\omega\}}+(\omega-\varepsilon_{(t,y)})1_{\{(t,y)\in supp\omega\}}. Denote {\mathbb{P}}_{N}:={\mathbb{P}}(d\omega)N_{\omega}(dt,dy) . Then it is well known (see [ 4 , Lemma 4.2] ) that the map (\omega,(t,y))\mapsto(\varepsilon_{(t,y)}^{+}\omega,(t,y)) sends {\mathbb{P}}_{N} -negligible sets to {\mathbb{P}}\times\mu -negligible ones, and the

Read the complete text