Kernel smoothing in nonparametric autoregressive schemes offers a powerful tool in modelling time series. We show that the bootstrap can be used for estimating the distribution of kernel smoothers. This can be done by mimicking the stochastic nature of the whole process in the bootstrap resample or by generating a simple regression model. Consistency of these bootstrap procedures will be shown.
Publié le : 2002-02-15
Classification:
bandwidth selection,
bootstrap,
kernel estimates,
local polynomial estimates,
nonparametric heteroscedastic autoregression,
nonparametric time series
@article{1078951087,
author = {Franke, J\"urgen and Kreiss, Jens-Peter and Mammen, Enno},
title = {Bootstrap of kernel smoothing in nonlinear time series},
journal = {Bernoulli},
volume = {8},
number = {2},
year = {2002},
pages = { 1-37},
language = {en},
url = {http://dml.mathdoc.fr/item/1078951087}
}
Franke, Jürgen; Kreiss, Jens-Peter; Mammen, Enno. Bootstrap of kernel smoothing in nonlinear time series. Bernoulli, Tome 8 (2002) no. 2, pp. 1-37. http://gdmltest.u-ga.fr/item/1078951087/