KERNEL ESTIMATION FOR ADDITIVE-MODELS UNDER DEPENDENCE

Authors
BAEK J WEHRLY TE
Citation
J. Baek et Te. Wehrly, KERNEL ESTIMATION FOR ADDITIVE-MODELS UNDER DEPENDENCE, Stochastic processes and their applications, 47(1), 1993, pp. 95-112
Citations number
25
Categorie Soggetti
Statistic & Probability","Statistic & Probability
Journal title
Stochastic processes and their applications → ACNP
ISSN journal
03044149
Volume
47
Issue
1
Year of publication
1993
Pages
95 - 112
Database
ISI
SICI code
0304-4149(1993)47:1<95:KEFAUD>2.0.ZU;2-#
Abstract
Nonparametric estimation of the conditional mean function for additive models is investigated in cases where the observed data are dependent . We use an additive kernel estimator which is a sum of Nadaraya-Watso n estimators. Under a strong mixing condition, the kernel estimator is shown to be asymptotically normal and to achieve the univariate optim al rate of convergence in mean squared error.