Statistics and Its Interface

Volume 9 (2016)

Number 3

Optimal bandwidth selection for semi-recursive kernel regression estimators

Pages: 375 – 388

DOI: http://dx.doi.org/10.4310/SII.2016.v9.n3.a11

Author

Yousri Slaoui (Laboratoire de Mathématiques et Application, Futuroscope, Chasseneuil, France)

Abstract

In this paper we propose an automatic selection of the bandwidth of the semi-recursive kernel estimators of a regression function defined by the stochastic approximation algorithm. We showed that, using the selected bandwidth and some special stepsizes, the proposed semi-recursive estimators will be very competitive to the nonrecursive one in terms of estimation error but much better in terms of computational costs. We corroborated these theoretical results through simulation study and a real dataset.

Keywords

nonparametric regression, stochastic approximation algorithm, smoothing, curve fitting

2010 Mathematics Subject Classification

Primary 62G08, 62L20. Secondary 65D10.

Full Text (PDF format)

Published 27 January 2016