Abstract
We propose a new numerical scheme for Backward Stochastic Differential Equations (BSDEs) based on branching processes. We approximate an arbitrary (Lipschitz) driver by local polynomials and then use a Picard iteration scheme. Each step of the Picard iteration can be solved by using a representation in terms of branching diffusion systems, thus avoiding the need for a fine time discretization. In contrast to the previous literature on the numerical resolution of BSDEs based on branching processes, we prove the convergence of our numerical scheme without limitation on the time horizon. Numerical simulations are provided to illustrate the performance of the algorithm.
Funding statement: Bruno Bouchard and Xavier Warin acknowledge the financial support of ANR project CAESARS (ANR-15-CE05-0024). Xiaolu Tan acknowledges the financial support of the ERC 321111 Rofirm, the ANR Isotace (ANR-12-MONU-0013), and the Chairs Financial Risks (Risk Foundation, sponsored by Société Générale) and Finance and Sustainable Development (IEF sponsored by EDF and CA). Xavier Warin is supported in part by the ANR project CAESARS (ANR-15-CE05-0024). This work has benefited from the financial support of the Initiative de Recherche “Méthodes non-linéaires pour la gestion des risques financiers” sponsored by AXA Research Fund.
A Appendix
A.1 Technical lemmas
Lemma A.1.
The ordinary differential equation
Moreover, it is bounded on
Consequently, one has, for all
Proof.
(i) We first claim that
Then, for every
This means that
(ii) Let us now prove (A.4).
Notice that
By direct computation, the left-hand side of (A.5) equals
When
(iii) We now prove (A.3).
Recall that
Since
It follows by Fatou’s Lemma that
as desired. ∎
For completeness, we provide here the proof the representation formula of Proposition 2.5 and of the technical lemma that was used in the proof of Proposition 2.8.
Proposition A.2.
The representation formula of Proposition 2.5 holds.
Proof.
We only provide the proof on
First, Lemma A.1 shows that the random variable
Next, set
where
satisfies
by (2.12). On the other hand, (2.11) and (2.12) imply
and
Combining the above implies that
and the required result follows by induction. ∎
Lemma A.3.
Let
Proof.
It suffices to observe that
and to proceed by induction. ∎
Proposition A.4.
Let
Then
Proof.
We have
The required result then follows from a simple induction. ∎
A.2 More on the error analysis for the abstract numerical approximation
The regression error
Since
for all
Proposition A.5.
Suppose that
Proof.
For ease of notations, we provide the proof for
(i) Let
and then
Since
By plugging the above estimates into (A.8), it follows that
(ii) For the Hölder property of
where the last inequality follows from the Lipschitz property of
References
[1] V. Bally and P. Pages, Error analysis of the quantization algorithm for obstacle problems, Stochastic Process. Appl. 106 (2003), no. 1, 1–40. 10.1016/S0304-4149(03)00026-7Search in Google Scholar
[2] C. Bender and R. Denk, A forward scheme for backward SDEs, Stochastic Process. Appl. 117 (2007), no. 12, 1793–1812. 10.1016/j.spa.2007.03.005Search in Google Scholar
[3] M. Bossy, N. Champagnat, H. Leman, S. Maire, L. Violeau and M. Yvinec, Monte Carlo methods for linear and non-linear Poisson–Boltzmann equation, ESAIM Proc. Surv. 48 (2015), 420–446. 10.1051/proc/201448020Search in Google Scholar
[4] B. Bouchard and N. Touzi, Discrete-time approximation and Monte-Carlo simulation of backward stochastic differential equations, Stochastic Process. Appl. 111 (2004), no. 2, 175–206. 10.1016/j.spa.2004.01.001Search in Google Scholar
[5] B. Bouchard and X. Warin, Monte-Carlo valuation of American options: Facts and new algorithms to improve existing methods, Numerical Methods in Finance, Springer Proc. Math. 12, Springer, Heidelberg (2012), 215–255. 10.1007/978-3-642-25746-9_7Search in Google Scholar
[6] H.-J. Bungartz, Concepts for higher order finite elements on sparse grids, Proceedings of the Third International Conference on Spectral and High Order Methods, University of Houston, Houston (1996), 159–170. Search in Google Scholar
[7] H.-J. Bungartz and M. Griebel, Sparse grids, Acta Numer. 13 (2004), 147–269. 10.1017/S0962492904000182Search in Google Scholar
[8] C. de Boor, A Practical Guide to Splines, Appl. Math. Sci. 27, Springer, New York, 1978. 10.1007/978-1-4612-6333-3Search in Google Scholar
[9] N. El Karoui, S. Peng and M. C. Quenez, Backward stochastic differential equations in finance, Math. Finance 7 (1997), no. 1, 1–71. 10.1111/1467-9965.00022Search in Google Scholar
[10] S. M. Ermakov, V. V. Nekrutkin and A. S. Sipin, Random Processes for Classical Equations of Mathematical Physics, Math. Appl. (Soviet Series) 34, Kluwer Academic Publishers, Dordrecht, 1989. 10.1007/978-94-009-2243-3Search in Google Scholar
[11] T. Gerstner and M. Griebel, Dimension-adaptive tensor-product quadrature, Computing 71 (2003), no. 1, 65–87. 10.1007/s00607-003-0015-5Search in Google Scholar
[12] H. Gevret, J. Lelong and X. Warin, The StOpt library documentation, preprint (2016), https://hal.archives-ouvertes.fr/hal-01361291. Search in Google Scholar
[13] H. Gevret, J. Lelong and X. Warin, The StOpt library, preprint, https://gitlab.com/stochastic-control/StOpt. Search in Google Scholar
[14] E. Gobet, J.-P. Lemor and X. Warin, A regression-based Monte Carlo method to solve backward stochastic differential equations, Ann. Appl. Probab. 15 (2005), no. 3, 2172–2202. 10.1214/105051605000000412Search in Google Scholar
[15] M. Griebel, Adaptive sparse grid multilevel methods for elliptic PDEs based on finite differences, Computing 61 (1998), no. 2, 151–179. 10.1007/BF02684411Search in Google Scholar
[16] P. Henry-Labordère, Cutting CVA’s complexity, Risk 25 (2012), no. 7, 67–67. Search in Google Scholar
[17] P. Henry-Labordere, N. Oudjane, X. Tan, N. Touzi and X. Warin, Branching diffusion representation of semilinear PDEs and Monte Carlo approximation, preprint (2016), https://arxiv.org/abs/1603.01727. 10.1214/17-AIHP880Search in Google Scholar
[18] P. Henry-Labordère, X. Tan and N. Touzi, A numerical algorithm for a class of BSDEs via the branching process, Stochastic Process. Appl. 124 (2014), no. 2, 1112–1140. 10.1016/j.spa.2013.10.005Search in Google Scholar
[19] J. Ma and J. Zhang, Representation theorems for backward stochastic differential equations, Ann. Appl. Probab. 12 (2002), no. 4, 1390–1418. 10.1214/aoap/1037125868Search in Google Scholar
[20] H. P. McKean, Application of Brownian motion to the equation of Kolmogorov–Petrovskii–Piskunov, Comm. Pure Appl. Math. 28 (1975), no. 3, 323–331. 10.1002/cpa.3160280302Search in Google Scholar
[21] E. Pardoux and S. G. Peng, Adapted solution of a backward stochastic differential equation, Systems Control Lett. 14 (1990), no. 1, 55–61. 10.1016/0167-6911(90)90082-6Search in Google Scholar
[22] A. Rasulov, G. Raimova and M. Mascagni, Monte Carlo solution of Cauchy problem for a nonlinear parabolic equation, Math. Comput. Simulation 80 (2010), no. 6, 1118–1123. 10.1016/j.matcom.2009.12.009Search in Google Scholar
[23] A. V. Skorokhod, Branching diffusion processes, Theory Probab. Appl. 9 (1964), no. 3, 445–449. 10.1137/1109059Search in Google Scholar
[24] X. Warin, Adaptive sparse grids for time dependent Hamilton–Jacobi–Bellman equations in stochastic control, preprint (2014), https://arxiv.org/abs/1408.4267. Search in Google Scholar
[25] X. Warin, Some non-monotone schemes for time dependent Hamilton–Jacobi–Bellman equations in stochastic control, J. Sci. Comput. 66 (2016), no. 3, 1122–1147. 10.1007/s10915-015-0057-9Search in Google Scholar
[26] S. Watanabe, On the branching process for Brownian particles with an absorbing boundary, J. Math. Kyoto Univ. 4 (1965), 385–398. 10.1215/kjm/1250524667Search in Google Scholar
© 2017 Walter de Gruyter GmbH, Berlin/Boston