Let X be a one dimensional positive recurrent diffusion continuously observed on [0,t] . We consider a non parametric estimator of the drift function on a given interval. Our estimator, obtained using a penalized least square approach, belongs to a finite dimensional functional space, whose dimension is selected according to the data. The non-asymptotic risk-bound reaches the minimax optimal rate of convergence when t → ∞. The main point of our work is that we do not suppose the process to be in stationary regime neither to be exponentially β-mixing. This is possible thanks to the use of a new polynomial inequality in the ergodic theorem [E. Löcherbach, D. Loukianova and O. Loukianov, Ann. Inst. H. Poincaré Probab. Statist. 47 (2011) 425-449].
Mots clés : diffusion process, adaptive estimation, regeneration method, mean square estimator, model selection, deviation inequalities
@article{PS_2011__15__197_0, author = {L\"ocherbach, Eva and Loukianova, Dasha and Loukianov, Oleg}, title = {Penalized nonparametric drift estimation for a continuously observed one-dimensional diffusion process}, journal = {ESAIM: Probability and Statistics}, pages = {197--216}, publisher = {EDP-Sciences}, volume = {15}, year = {2011}, doi = {10.1051/ps/2009016}, mrnumber = {2870512}, language = {en}, url = {http://www.numdam.org/articles/10.1051/ps/2009016/} }
TY - JOUR AU - Löcherbach, Eva AU - Loukianova, Dasha AU - Loukianov, Oleg TI - Penalized nonparametric drift estimation for a continuously observed one-dimensional diffusion process JO - ESAIM: Probability and Statistics PY - 2011 SP - 197 EP - 216 VL - 15 PB - EDP-Sciences UR - http://www.numdam.org/articles/10.1051/ps/2009016/ DO - 10.1051/ps/2009016 LA - en ID - PS_2011__15__197_0 ER -
%0 Journal Article %A Löcherbach, Eva %A Loukianova, Dasha %A Loukianov, Oleg %T Penalized nonparametric drift estimation for a continuously observed one-dimensional diffusion process %J ESAIM: Probability and Statistics %D 2011 %P 197-216 %V 15 %I EDP-Sciences %U http://www.numdam.org/articles/10.1051/ps/2009016/ %R 10.1051/ps/2009016 %G en %F PS_2011__15__197_0
Löcherbach, Eva; Loukianova, Dasha; Loukianov, Oleg. Penalized nonparametric drift estimation for a continuously observed one-dimensional diffusion process. ESAIM: Probability and Statistics, Tome 15 (2011), pp. 197-216. doi : 10.1051/ps/2009016. http://www.numdam.org/articles/10.1051/ps/2009016/
[1] Nonparametric identification for diffusion processes. SIAM J. Control. Optim. 16 (1978) 380-395. | MR | Zbl
,[2] Adaptive estimation in autoregression or β-mixing regression via model selection. Ann. Stat. 29 (2001) 839-875. | MR | Zbl
, and ,[3] Model selection for (auto-)regression with dependent data. ESAIM: P&S 5 (2001) 33-49. | Numdam | MR | Zbl
, and ,[4] Risks bounds for model selection via penalization. Prob. Th. Rel. Fields 113 (1999) 301-413. | MR | Zbl
, and ,[5] Penalized nonparametric mean square estimation of the coefficients of diffusion processes. Bernoulli 13 (2007) 514-543. | MR | Zbl
, and ,[6] Sharp adaptive estimation of the drift function for ergodic diffusions. Ann. Stat. 33 (2005) 507-2528. | MR | Zbl
,[7] Asymptotically efficient trend coefficient estimation for ergodic diffusion. Math. Meth. Stat. 11 (2002) 402-427. | MR
and ,[8] Dynamics adaptive estimation of a scalar diffusion. Prpublication PMA-762, Univ. Paris 6 (2002). Available at www.proba.jussieu.fr/mathdoc/preprints/. Mathematical Reviews (MathSciNet): MR1895888 Project Euclid: euclid.bj/1078866865. | MR
, and ,[9] Sequential nonparametric adaptive estimation of the drift coefficient in the diffusion processes. Math. Meth. Stat. 10 (2001) 316-330. | MR | Zbl
and ,[10] Adaptive estimation in diffusion processes. Stochastic Processes Appl. 79 (1999) 135-163. | MR | Zbl
,[11] Statistical inference for ergodic diffusion processes. Springer Series in Statistics. London: Springer (2004). | MR | Zbl
,[12] One problem of adaptive estimation in Gaussian white noise. Theory Probab. Appl. 35 (1999) 459-470. | MR | Zbl
,[13] Constructive approximation: advanced problems. Grundlehren der Mathematischen Wissenschaften 304. Berlin: Springer (1996). | MR | Zbl
, and ,[14] Uniform deterministic equivalent of additive functionals and non-parametric drift estimation for one-dimensional recurrent diffusions. Ann. Inst. Henri Poincaré 44 (2008) 771-786. | Numdam | MR | Zbl
and ,[15] On Nummelin splitting for continuous time Harris recurrent Markov processes and application to kernel estimation for multi-dimensional diffusions. Stoch. Proc. Appl. 118 (2008) 301-1321. | MR | Zbl
and ,[16] Polynomial bounds in the Ergodic theorem for one-dimensional diffusions and integrability of hitting times. Ann. Inst. H. Poincaré Probab. Statist. 47 (2011) 425-449. | Numdam | MR | Zbl
, and ,[17] Nonparametric estimation of the drift coefficient in the diffusion equation. Math. Operationsforsch. Statist., Ser. Statistics 1 (1981) 61-73. | MR | Zbl
,[18] Statistical Inference for Diffusion Type Processes. London: Edward Arnold. MR1717690 (1999) | MR | Zbl
,[19] Continuous martingales and Brownian motion. 3rd ed. Grundlehren der Mathematischen Wissenschaften 293. Berlin: Springer (2005). | Zbl
and ,[20] Adaptive drift estimation for nonparametric diffusion model. Ann. Stat. 28 (2000) 815-836. | MR | Zbl
,[21] On polynomial mixing bounds for stochastic differential equations. Stoch. Proc. Appl. 70 (1997) 115-127. | MR | Zbl
,Cité par Sources :