Nonparametric estimation of the density of regression errors

Rawane Samb

doi:10.1016/j.crma.2011.10.017

Probability Theory/Statistics

Nonparametric estimation of the density of regression errors

Presented by: Paul Deheuvels

Rawane Samb ¹

¹ ISBA, Université Catholique de Louvain, 20, voie du Roman pays, B-1348 Louvain-la-Neuve, Belgium

Comptes Rendus. Mathématique, Volume 349 (2011) no. 23-24, pp. 1281-1285.

Abstract (OV)
Résumé

Consider the nonparametric regression model $Y = m (X) + ε$ , where the function m is smooth but unknown, and ε is independent of X. An estimator of the density of the error term ε is proposed and its weak consistency is obtained. The strategy used here is based on the kernel estimation of the residuals. Our contribution is twofold. First, we evaluate the impact of the estimation of the regression function m on the error density estimator. Secondly, the optimal choices of the first and second-step bandwidths used for estimating the regression function and the error density respectively, are proposed. Further, we investigate the asymptotic normality of the error density estimator and its rate-optimality.

Nous présentons un estimateur nonparamétrique de la densité de lʼerreur dans le modèle de régression $Y = m (X) + ε$ , où la fonction m est lisse mais inconnue, et le terme dʼerreur ε est indépendant de X. Lʼestimateur proposé est basé sur une estimation nonparamétrique des résidus, et sa consistance faible est obtenue. Notre contribution se situe à deux niveaux. Dʼabord, nous évaluons lʼimpact de lʼestimation de la fonction de régression sur lʼestimateur final de la densité de lʼerreur. Ensuite, nous proposons les choix optimaux des fenêtres de première et de deuxième étape utilisées respectivement pour les estimations de m et de la densité des résidus. Nous étudions également la normalié asymptotique de lʼestimateur de la densité de lʼerreur et sa vitesse de convergence.

Received: 2011-08-19
Accepted: 2011-10-19
Published online: 2011-11-09

DOI: 10.1016/j.crma.2011.10.017

Author's affiliations:

Rawane Samb ¹

¹ ISBA, Université Catholique de Louvain, 20, voie du Roman pays, B-1348 Louvain-la-Neuve, Belgium

@article{CRMATH_2011__349_23-24_1281_0,
     author = {Rawane Samb},
     title = {Nonparametric estimation of the density of regression errors},
     journal = {Comptes Rendus. Math\'ematique},
     pages = {1281--1285},
     publisher = {Elsevier},
     volume = {349},
     number = {23-24},
     year = {2011},
     doi = {10.1016/j.crma.2011.10.017},
     language = {en},
}

TY  - JOUR
AU  - Rawane Samb
TI  - Nonparametric estimation of the density of regression errors
JO  - Comptes Rendus. Mathématique
PY  - 2011
SP  - 1281
EP  - 1285
VL  - 349
IS  - 23-24
PB  - Elsevier
DO  - 10.1016/j.crma.2011.10.017
LA  - en
ID  - CRMATH_2011__349_23-24_1281_0
ER  -

%0 Journal Article
%A Rawane Samb
%T Nonparametric estimation of the density of regression errors
%J Comptes Rendus. Mathématique
%D 2011
%P 1281-1285
%V 349
%N 23-24
%I Elsevier
%R 10.1016/j.crma.2011.10.017
%G en
%F CRMATH_2011__349_23-24_1281_0

Rawane Samb. Nonparametric estimation of the density of regression errors. Comptes Rendus. Mathématique, Volume 349 (2011) no. 23-24, pp. 1281-1285. doi : 10.1016/j.crma.2011.10.017. https://comptes-rendus.academie-sciences.fr/mathematique/articles/10.1016/j.crma.2011.10.017/

References
Cited by

[1] M.G. Akritas; I. Van Keilegom Non-parametric estimation of the residual distribution, Scand. J. Stat., Volume 28 (2001), pp. 549-567

[2] F. Cheng Asymptotic distributions of error density and distribution function estimators in nonparametric regression, J. Statist. Plann. Inference, Volume 128 (2005), pp. 327-349

[3] P. Deheuvels; D.M. Mason General asymptotic confidence bands based on kernel-type function estimators, Stat. Inference Stoch. Process., Volume 7 (2004), pp. 225-277

[4] S. Efromovich Estimation of the density of the regression errors, Ann. Statist., Volume 33 (2005), pp. 2194-2227

[5] U.U. Müller; A. Schick; W. Wefelmeyer Estimating the error distribution in nonparametric regression with multivariate covariates, Statist. Probab. Lett., Volume 79 (2009), pp. 957-964

[6] E.A. Nadaraya On a regression estimate, Teor. Verojatnost. i Primenen., Volume 9 (1964), pp. 157-159

[7] N. Neumeyer; I. Van Keilegom Estimating the error distribution in nonparametric multiple regression with applications to model testing, J. Multivariate Anal., Volume 101 (2010), pp. 1067-1078

[8] M.S. Pinsker Optimal filtering of a square integrable signal in Gaussian white noise, Probl. Inf. Transm., Volume 16 (1980), pp. 52-68

[9] S. Plancade Nonparametric estimation of the density of the regression noise, C. R. Acad. Sci. Paris, Ser. I, Volume 346 (2008), pp. 461-466

[10] W.S. Scott Multivariate Density Estimation, Wiley, 1992

[11] B.W. Silverman Density Estimation for Statistics and Data Analysis, Monographs on Statistics and Applied Probability, vol. 26, Chapman and Hall, London, 1986

[12] M.P. Wand; M.C. Jones Kernel Smoothing, Chapman & Hall/CRC, 1995

[13] G.S. Watson Smooth regression analysis, Sankhyā Ser. A, Volume 26 (1964), pp. 359-372

Cited by Sources:

Comments - Policy