Abstract
In this paper, we investigate the problem of estimating the regression function in models with correlated observations. The data are obtained from several experimental units, each of them forms a time series. Using the properties of the reproducing kernel Hilbert spaces, we construct a new estimator based on the inverse of the autocovariance matrix of the observations. We give the asymptotic expressions of its bias and its variance. In addition, we give a theoretical comparison between this new estimator and the popular one proposed by Gasser and Müller, we show that the proposed estimator has an asymptotically smaller variance than the classical one. Finally, we conduct a simulation study to investigate the performance and the robustness of the proposed estimator and to compare it to the Gasser and Müller’s estimator in a finite sample set.
Similar content being viewed by others
Notes
The kernel K satisfies: \(\int _{-1}^1 K(t)\hbox {d}t=1,\) \(\int _{-1}^1 t K(t)\hbox {d}t=0\) and \(\int _{-1}^1 t^2 K(t)\hbox {d}t <+ \infty .\)
References
Altman, N. S. (1990). Kernel smoothing of data with correlated errors. American Statistical Association, 85, 749–759.
Azzalini, A. (1984). Estimation and hypothesis testing for collections of autoregressive time series. Biometrika, 71, 85–90.
Belouni, M., Benhenni, K. (2015). Optimal and robust designs for estimating the concentration curve and the AUC. Scandinavian Journal of Statistics: Theory and Application, 42, 453–470.
Benedetti, J. (1977). On the nonparametric estimation of the regression function. Journal of the Royal Statistical Society, 39, 248–253.
Benelmadani, D. (2019a). Contribution à la Régression non Paramétrique avec un Processus Erreur d’Autocovariance Générale et Application en Pharmacocinétique, Ph.D. 2019, Grenoble Alpes University, France.
Benelmadani, D., Benhenni, K., Louhichi, S. (2019b). Trapezoidal rule and sampling designs for the nonparametric estimation of the regression function in models with correlated errors. arXiv:1806.04896.
Benhenni, K., Cambanis, S. (1992). Sampling designs for estimating integrals of stochastic processes. The Annals of Statistics, 20, 161–194.
Benhenni, K., Rachdi, M. (2007). Nonparametric estimation of average growth curve with general nonstationary error process. Communications in Statistics: Theory and Methods, 36, 1137–1186.
Blanke, D., Bosq, D. (2008). Regression estimation and prediction in continuous time. Journal of the Japan Statistical Society (Nihon Tôkei Gakkai Kaihô), 38, 15–26.
Cheng, K. F., Lin, P. E. (1981). Nonparametric estimation of the regression function. Z. Wahrscheinlichkeitstheorie verw. Gebiete, 57, 223–233.
Chiu, S. T. (1989). Bandwidth selection for kernel estimation with correlated noise. Statistics and Probability Letters, 8, 347–354.
Dette, H., Pepelyshev, A., Zhigljavsky, A. (2016). Best linear unbiased estimators in continuous time regression models. arXiv:1611.09804.
Didi, S., Louani, D. (2013). Asymptotic results for the regression function estimate on continuous time stationary and ergodic data. Journal of Statistics and Risk Modelling, 31(2), 129–150.
Ferreira, E., Núǹez-Antón, V., Rodríguez-Póo, J. (1997). Kernel regression estimates of growth curves using nonstationary correlated errors. Statistics and Probability Letters, 34, 413–423.
Gasser, T., Müller, H. G. (1979). Kernel estimation of regression functions. Lecture Notes in Mathematics, 757, 23–68.
Hart, J. D. (1991). Kernel regression estimation with time series errors. Royal Statistical Society B, 53, 173–187.
Hart, J. D. (1994). Automated kernel smoothing of dependent data by using time series cross validation. Royal Statistical Society B, 56, 529–542.
Hart, J. D., Wherly, T. E. (1986). Kernel regression estimation using repeated measurements data. Journal of the American Statistical Association, 81, 1080–1088.
Müller, G. H. (1984). Optimal designs for nonparametric kernel regression. Statistics and Probability Letters, 2, 285–290.
Parzen, E. (1959). Statistical inference on time series by hilbert space methods, p. 23. Technical Report, Department of Statistics, Stanford University, Stanford, CA.
Priestly, M. B., Chao, M. T. (1972). Nonparametric function fitting. Journal of Royal Statistical Society, 34, 384–392.
Ramsay, J. O., Silverman, B. W. (2005). Functional data analysis, Springer series in statistics, New York: Springer.
Rice, J. (1984). Bandwidth choice for nonparametric regression. The Annals of Statistics, 12, 1215–1230.
Sacks, J., Ylvisaker, D. (1966). Designs for regression problems with correlated errors. The Annals of Mathematical Statistics, 37, 66–89.
Sacks, J., Ylvisaker, D. (1968). Designs for regression problems with correlated errors: Many parameters. The Annals of Mathematical Statistics, 39, 69–79.
Sacks, J., Ylvisaker, D. (1970). Designs for regression problems with correlated errors III. The Annals of Mathematical Statistics, 41, 2057–2074.
Schwartz, L. (1964). Sous Espace Hilbertiens d’Espace Vectoriels Topologiques et Noyaux Associés (Noyaux Reproduisant). Journal d’Analyse Mathématique, 13, 115–256.
Su, Y., Cambanis, S. (1993). Sampling designs for estimation of a random process. Stochastic Processes and their Applications, 46, 47–89.
Xiang, Y., Gubian, S., Suomela, B., Hoeng, J. (2013). Generalized simulated annealing for efficient global optimization: The GenSA package for R. The R Journal, 5, 13–28.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Electronic supplementary material
Below is the link to the electronic supplementary material.
About this article
Cite this article
Benelmadani, D., Benhenni, K. & Louhichi, S. The reproducing kernel Hilbert space approach in nonparametric regression problems with correlated observations. Ann Inst Stat Math 72, 1479–1500 (2020). https://doi.org/10.1007/s10463-019-00733-3
Received:
Revised:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10463-019-00733-3