This paper investigates the (conditional) quasi-likelihood ratio test for the threshold in MA models. Under the hypothesis of no threshold, it is shown that the test statistic converges weakly to a function of the centred Gaussian process. Under local alternatives, it is shown that this test has nontrivial asymptotic power. The results are based on a new weak convergence of a linear marked empirical process, which is independently of interest. This paper also gives an invertible expansion of the threshold MA models.
Publié le : 2005-12-14
Classification:
Invertibility,
likelihood ratio test,
MA model,
marked empirical process,
threshold MA model,
62F05,
62M10,
60G10
@article{1140191665,
author = {Ling, Shiqing and Tong, Howell},
title = {Testing for a linear MA model against threshold MA models},
journal = {Ann. Statist.},
volume = {33},
number = {1},
year = {2005},
pages = { 2529-2552},
language = {en},
url = {http://dml.mathdoc.fr/item/1140191665}
}
Ling, Shiqing; Tong, Howell. Testing for a linear MA model against threshold MA models. Ann. Statist., Tome 33 (2005) no. 1, pp. 2529-2552. http://gdmltest.u-ga.fr/item/1140191665/