Two consistent estimators for the skew Brownian motion
ESAIM: Probability and Statistics, Tome 23 (2019), pp. 567-583.

The skew Brownian motion (SBm) is of primary importance in modeling diffusion in media with interfaces which arise in many domains ranging from population ecology to geophysics and finance. We show that the maximum likelihood procedure estimates consistently the parameter of an SBm observed at discrete times. The difficulties arise because the observed process is only null recurrent and has a singular distribution with respect to the one of the Brownian motion. Finally, using the idea of the expectation–maximization algorithm, we show that the maximum likelihood estimator can be naturally interpreted as the expected total number of positive excursions divided by the expected number of excursions given the observations. The theoretical results are illustrated by numerical simulations.

DOI : 10.1051/ps/2018018
Classification : 60H10, 62F12
Mots-clés : Skew Brownian motion, maximum likelihood estimator (MLE) null recurrent process, expectation–maximization (EM) algorithm, excursion theory
Lejay, Antoine 1 ; Mordecki, Ernesto 1 ; Torres, Soledad 1

1
@article{PS_2019__23__567_0,
     author = {Lejay, Antoine and Mordecki, Ernesto and Torres, Soledad},
     title = {Two consistent estimators for the skew {Brownian} motion},
     journal = {ESAIM: Probability and Statistics},
     pages = {567--583},
     publisher = {EDP-Sciences},
     volume = {23},
     year = {2019},
     doi = {10.1051/ps/2018018},
     mrnumber = {3991913},
     zbl = {1506.60057},
     language = {en},
     url = {http://www.numdam.org/articles/10.1051/ps/2018018/}
}
TY  - JOUR
AU  - Lejay, Antoine
AU  - Mordecki, Ernesto
AU  - Torres, Soledad
TI  - Two consistent estimators for the skew Brownian motion
JO  - ESAIM: Probability and Statistics
PY  - 2019
SP  - 567
EP  - 583
VL  - 23
PB  - EDP-Sciences
UR  - http://www.numdam.org/articles/10.1051/ps/2018018/
DO  - 10.1051/ps/2018018
LA  - en
ID  - PS_2019__23__567_0
ER  - 
%0 Journal Article
%A Lejay, Antoine
%A Mordecki, Ernesto
%A Torres, Soledad
%T Two consistent estimators for the skew Brownian motion
%J ESAIM: Probability and Statistics
%D 2019
%P 567-583
%V 23
%I EDP-Sciences
%U http://www.numdam.org/articles/10.1051/ps/2018018/
%R 10.1051/ps/2018018
%G en
%F PS_2019__23__567_0
Lejay, Antoine; Mordecki, Ernesto; Torres, Soledad. Two consistent estimators for the skew Brownian motion. ESAIM: Probability and Statistics, Tome 23 (2019), pp. 567-583. doi : 10.1051/ps/2018018. http://www.numdam.org/articles/10.1051/ps/2018018/

[1] L.H.R. Alvarez and P. Salminen, Timing in the presence of directional predictability: Optimal stopping of skew brownian motion. Math. Methods Oper. Res. 86 (2017) 377–400. | DOI | MR | Zbl

[2] T.A. Appuhamillage, V.A. Bokil, E. Thomann, E. Waymire and B.D. Wood, Occupation and local times for skew brownian motion with application to dispersion accross an interface. Ann. Appl. Probab. 21 (2011) 183–214. | MR | Zbl

[3] O. Bardou and M. Martinez, Statistical estimation for reflected skew processes. Stat. Inference Stoch. Process. 13 (2010) 231–248. | DOI | MR | Zbl

[4] M. Bossy, N. Champagnat, S. Maire and D. Talay, Probabilistic interpretation and random walk on spheres algorithms for the Poisson-Boltzmann equation in molecular dynamics. Math. Model. Numer. Anal. 44 (2010) 997–1048. | DOI | Numdam | MR | Zbl

[5] M. Decamps, M. Goovaerts and W. Schoutens, Self exciting threshold interest rates models. Int. J. Theor. Appl. Finance 9 (2006) 1093–1122. | DOI | MR | Zbl

[6] A.P. Dempster, N.M. Laird and D.B. Rubin, Maximum likelihood from incomplete data via the EM algorithm. J. R. Stat. Soc. Ser. B (Methodol.) 39 (1977) 1–38. | MR | Zbl

[7] D. Florens, Estimation of the diffusion coefficient from crossings. Stat. Inference Stoch. Process. 1 (1998) 175–195. | DOI | MR | Zbl

[8] J.M. Harrison and L.A. Shepp, On skew Brownian motion. Ann. Probab. 9 (1981) 309–313. | DOI | MR | Zbl

[9] R. Höpfner and E. Löcherbach, Limit theorems for null recurrent Markov processes. Mem. Am. Math. Soc. 161 (2003). | MR | Zbl

[10] K. Itô and H.P. Mckean, Jr., Diffusion Processes and Their Sample Paths, 2nd edn. Springer-Verlag, Berlin, New York (1974). | MR | Zbl

[11] J. Jacod, Une généralisation des semimartingales: les processus admettant un processus à accroissements indépendants tangent. in Seminaire de probabilités XVIII. Vol. 1059 of Lecture Notes in Mathematics. Springer, Berlin (1984) 91–118. | DOI | Numdam | MR | Zbl

[12] J. Jacod, Rates of convergence to the local time of a diffusion. Ann. Inst. Henri Poincaré Probab. Stat. 34 (1998) 505–544. | DOI | Numdam | MR | Zbl

[13] J. Keilson and J.A. Wellner, Oscillating Brownian motion. J. Appl. Probab. 15 (1978) 300–310. | DOI | MR | Zbl

[14] J.-F. Le Gall One-dimensional stochastic differential equations involving the local times of the unknown process, in Stochastic Analysis and Applications, Swansea, 1983. Vol. 1095 of Lecture Notes in Mathematics. Springer, Berlin, Heidelberg (1984) 51–82. | MR | Zbl

[15] J.-F. Le Gall One-dimensional stochastic differential equations involving the local times of the unknown process, in Stochastic Analysis and Applications. Vol. 1095 of Lecture Notes in Mathematics. Springer Verlag, Berlin (1985) 51–82. | DOI | MR | Zbl

[16] A. LejayOn the constructions of the skew Brownian motion. Probab. Surv. 3 (2006) 413–466. | DOI | MR | Zbl

[17] A. LejayEstimation of the biais parameter of the Skew Random Walk and application to the Skew Brownian Motion. Stat. Inference Stoch. Process. 21 (2018) 539–551. | DOI | MR

[18] A. Lejay and G. Pichot, Simulating diffusion processes in discontinuous media: a numerical scheme with constant time steps. J. Comput. Phys. 231 (2012) 7299–7314. | DOI | MR | Zbl

[19] A. Lejay and G. Pichot, Simulating diffusion processes in discontinuous media: Benchmark tests. J. Comput. Phys. 314 (2016) 348–413. | DOI | MR | Zbl

[20] A. Lejay and P. Pigato, Statistical estimation of the oscillating brownian motion. Bernoulli 24 (2018) 3568–3602. | DOI | MR | Zbl

[21] A. Lejay and P. Pigato, A threshold model for local volatility: evidence of leverage and mean reversion effects on historical data. 2018. Preprint. | MR

[22] A. LejayE. Mordecki and S. Torres, Is a Brownian motion skew? Scand. J. Stat. 41 (2014) 346–364. | DOI | MR | Zbl

[23] D. Lépingle, Un schéma d’Euler pour équations différentielles stochastiques réfléchies. C. R. Acad. Sci. Paris, Sér. I Math. 316 (1993) 601–605. | MR | Zbl

[24] M. Martinez, Interprétations probabilistes d’opérateurs sous forme divergence et analyse de méthodes numériques associées. Ph.D. thesis, Université de Provence/INRIA Sophia-Antipolis (2004).

[25] G.J. Mclachlan and T. Krishnan, The EM Algorithm and Extensions, 2nd edn. Wiley Series in Probability and Statistics. Wiley-, NJ (2008). | MR

[26] O. Ovaskainen and S.J. Cornell, Biased movement at a boundary and conditional occupancy times for diffusion processes. J. Appl. Probab. 40 (2003) 557–580. | DOI | MR | Zbl

[27] N.I. Portenko, Diffusion processes with a generalized drift coefficient. Teor. Veroyatnost. i Primenen. 24 (1979) 62–77. | MR | Zbl

[28] D. Rossello, Arbitrage in skew Brownian motion models. Insurance Math. Econom. 50 (2012) 50–56. | DOI | MR | Zbl

[29] D. Spivakovsakaya, A.W. Heemink and E. Deleersnijder. The backward ito method for the lagrangian simulation of transport processes with large space variations of the diffusivity. Ocean Sci. 3 (2007) 525–535. | DOI

[30] D.J. Thomson, W.L. Physick and R.H. Maryon, Treatment of interfaces in random walk dispersion models. J. Appl. Meteorol. 36 (1997) 1284–1295. | DOI

[31] J.B. Walsh, A diffusion with discontinuous local time, in Temps Locaux. Vol. 52–53 of Astérisques. Société Mathématique de France, Marseille (1978) 37–45. | Numdam

[32] M. Zhang, Calculation of diffusive shock acceleration of charged particles by skew Brownian motion. Astrophys. J. 541 (2000) 428–435. | DOI

Cité par Sources :