Approximation of the solution of the backward stochastic differential equation. Small noise, large sample and high frequency cases

Yury A. Kutoyants

Trudy Matematicheskogo Instituta imeni V.A. Steklova, Stochastic calculus, martingales, and their applications, Tome 287 (2014), pp. 140-161.

Résumé

We present a review of some recently obtained results on estimation of the solution of a backward stochastic differential equation (BSDE) in the Markovian case. We suppose that the forward equation depends on some finite-dimensional unknown parameter. We consider the problem of estimating this parameter and then use the proposed estimator to estimate the solution of the BSDE. This last estimator is constructed with the help of the solution of the corresponding partial differential equation. We are interested in three observation models admitting a consistent estimation of the unknown parameter: small noise, large samples and unknown volatility. In the first two cases we have a continuous time observation, and the unknown parameter is in the drift coefficient. In the third case the volatility of the forward equation depends on the unknown parameter, and we have discrete time observations. The presented estimators of the solution of the BSDE in the three casesmentioned are asymptotically efficient.

Bibliographie

[1] Bismut J.-M., “Conjugate convex functions in optimal stochastic control”, J. Math. Anal. Appl., 44 (1973), 384–404 | DOI | MR

[2] Dohnal G., “On estimating the diffusion coefficient”, J. Appl. Probab., 24:1 (1987), 105–114 | DOI | MR | Zbl

[3] Freidlin M.I., Wentzell A.D. Random perturbations of dynamical systems. 2nd ed. New York: Springer, 1998. | MR

[4] Genon-Catalot V., Jacod J., “On the estimation of the diffusion coefficient for multi-dimensional diffusion processes”, Ann. Inst. Henri Poincaré. Probab. Stat., 29:1 (1993), 119–151 | MR | Zbl

[5] El Karoui N., Peng S., Quenez M.C., “Backward stochastic differential equations in finance”, Math. Finance., 7:1 (1997), 1–71 | DOI | MR | Zbl

[6] Kamatani K., Uchida M. Hybrid multi-step estimators for stochastic differential equations based on sampled data // Stat. Inference Stoch. Processes. 2014. doi: 10.1007/s11203-014-9107-4.

[7] Kutoyants Yu., Identification of dynamical systems with small noise, Kluwer, Dordrecht, 1994 | MR | Zbl

[8] Kutoyants Yu.A., Statistical inference for ergodic diffusion processes, Springer, London, 2004 | MR | Zbl

[9] Kutoyants Yu.A., Zhou L., “On approximation of the backward stochastic differential equation”, J. Stat. Plann. Inference., 150 (2014), 111–123, arXiv: 1305.3728 [math.ST] | DOI | MR | Zbl

[10] Levanony D., Shwartz A., Zeitouni O., “Recursive identification in continuous-time stochastic processes”, Stoch. Processes Appl., 49:2 (1994), 245–275 | DOI | MR | Zbl

[11] Liptser R.S., Shiryaev A.N. Statistics of random processes. 2nd ed. Berlin: Springer, 2001. V. 2.

[12] Ma J., Yong J., Forward–backward stochastic differential equations and their applications, (Lect. Notes Math.; V. 1702)., 1702, Springer, Berlin, 1999 | MR | Zbl

[13] Pardoux E., Peng S.G., “Adapted solution of a backward stochastic differential equation”, Syst. Control Lett., 14:1 (1990), 55–61 | DOI | MR | Zbl

[14] Pardoux E., Peng S. Backward stochastic differential equations and quasilinear parabolic differential equations // Stochastic partial differential equations and their applications. Berlin: Springer, 1992. P. 200–217. (Lect. Notes Control Inf. Sci.; V. 176). | MR

[15] Sørensen M. Estimating functions for diffusion-type processes // Statistical methods for stochastic differential equations / Ed. by M. Kessler, A. Lindner, M. Sørensen. Boca Raton, FL: CRC Press, 2009. P. 1–107. | MR

[16] Zhou L. Problèmes statistiques pour des EDS et les EDS rétrogrades: PhD Thesis. Le Mans: Univ. Maine, 2013.