This paper is concerned with model selection in time series analysis. An identification criterion is presented that is asymptotically equivalent to a Bayes decision rule. The discussion is conducted in the context of a general class of parametric time series models and consideration is given to the special case of order determination in autoregressive moving-average representations. Consistency of the criterion is proved.
Publié le : 1986-06-14
Classification:
Time series model,
power spectrum,
autoregressive moving-average representation,
Bayes decision rule,
model selection criterion,
consistency,
62M10,
62C10
@article{1176349935,
author = {Poskitt, D. S.},
title = {A Bayes Procedure for the Identification of Univariate Time Series Models},
journal = {Ann. Statist.},
volume = {14},
number = {2},
year = {1986},
pages = { 502-516},
language = {en},
url = {http://dml.mathdoc.fr/item/1176349935}
}
Poskitt, D. S. A Bayes Procedure for the Identification of Univariate Time Series Models. Ann. Statist., Tome 14 (1986) no. 2, pp. 502-516. http://gdmltest.u-ga.fr/item/1176349935/