For the autoregressive-moving average time series model, the normal theory procedure for setting confidence intervals for the error variance is not robust against nonnormality. This paper proposes three asymptotically robust techniques: they are a "standard-error" procedure, an analog of Box's simple data splitting technique, and the jackknife procedure. The large sample distribution of each of these techniques is derived.
Publié le : 1977-07-14
Classification:
Autoregressive-moving average,
discrete time series,
variance,
data splitting,
jackknife,
robust inference,
62M10,
62G35
@article{1176343893,
author = {Davis, William W.},
title = {Robust Interval Estimation of the Innovation Variance of an Arma Model},
journal = {Ann. Statist.},
volume = {5},
number = {1},
year = {1977},
pages = { 700-708},
language = {en},
url = {http://dml.mathdoc.fr/item/1176343893}
}
Davis, William W. Robust Interval Estimation of the Innovation Variance of an Arma Model. Ann. Statist., Tome 5 (1977) no. 1, pp. 700-708. http://gdmltest.u-ga.fr/item/1176343893/