Limit theorems for BSDE with local time applications to non-linear PDE (Q3148778)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 1804038
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | Limit theorems for BSDE with local time applications to non-linear PDE |
scientific article; zbMATH DE number 1804038 |
Statements
Limit theorems for BSDE with local time applications to non-linear PDE (English)
0 references
2 June 2003
0 references
backward stochastic differential equation
0 references
local time
0 references
comparison theorem
0 references
singular partial differential equations
0 references
0 references
0.7760935
0 references
0.7751739
0 references
0.77393657
0 references
0.7677969
0 references
0.76139796
0 references
0.7455463
0 references
0.7419121
0 references
0.74014246
0 references
In a former paper, \textit{A. Dermoune}, \textit{S. Hamadène} and \textit{Y. Ouknine} [ibid. 66, No. 1/2, 103-119 (1999; Zbl 0930.60044)] have studied the solution \((Y^\nu,Z^\nu)\) of the backward stochastic differential equation (BSDE) \[ dY^\nu_t=-d_t\Bigl( \int_RL^a_t (Y^\nu)\nu(da) \Bigr)+ Z^\nu_t dW_t, \quad Y^\nu_T= \xi\in L^2({\mathcal F}^W_T), \] where \(W\) is a Brownian motion, \(L^a (Y^\nu)\) denotes the symmetric local time of \(Y^\nu\) at level \(a\), and \(\nu (da)\) is a given signed measure. As main tool to study such BSDE the authors have used the transformation \(\nu\to F^\nu\) which has been introduced by \textit{D. W. Stroock} and \textit{M. Yor} [in: Séminaire de probabilités XV. Lect. Notes Math. 850, 590-603 (1981; Zbl 0456.60048)] [see also: \textit{J. F. Le Gall}, in: Stochastic analysis and applications. Lect. Notes Math. 1095, 51-82 (1984; Zbl 0551.60059)] in the context of SDE involving the local time of the solution process; the importance of this transformation comes from the fact that it allows to represent the solution of the BSDE in the following form: \(F^\nu (Y^\nu_t)= E[F^\nu(\xi) \mid{\mathcal F}_t^W]\). Now, in the present paper, given a sequence of signed measures \(\nu_n\), \(n\geq 1\), and a signed measure \(\nu\), that satisfy some appropriate boundedness assumptions and are such that \({d\over dx}F^{\nu_n}\) converges to \({d\over dx}F^\nu\) locally in \(L^2(R)\), the authors show with a straightforward calculus the convergence of \((Y^{\nu_n}, Z^{\nu_n})\) to \((Y^\nu,Z^\nu)\) in \(L^2\). This allows them to conclude that \((Y^\nu,Z^\nu)\) can be approximated by the solution \((Y,Z)\) of the BSDE \(dY_t=-f(Y^t) |Z_t|^2 dt+Z_tdW_t\). Using again the argument of the transformation \(\nu\to F^\nu\) the second part of the paper compares \(T_t^\nu\) and \(Y_t^\mu\) for signed measures \(\nu\leq\mu\) and applies these results on BSDE to associated PDEs.
0 references