Consistency of Akaike's Information Criterion for Infinite Variance Autoregressive Processes
Knight, Keith
Ann. Statist., Tome 17 (1989) no. 1, p. 824-840 / Harvested from Project Euclid
Suppose $\{X_n\}$ is a $p$th order autoregressive process with innovations in the domain of attraction of a stable law and the true order $p$ unknown. The estimate $\hat{p}$ of $p$ is chosen to minimize Akaike's information criterion over the integers $0, 1, \cdots, K$. It is shown that $\hat{p}$ is weakly consistent and the consistency is retained if $K \rightarrow \infty$ as $N \rightarrow \infty$ at a certain rate depending on the index of the stable law.
Publié le : 1989-06-14
Classification:  Akaike's information criterion,  autoregressive processes,  infinite variance,  stable law,  62M10,  62F12,  60G10
@article{1176347145,
     author = {Knight, Keith},
     title = {Consistency of Akaike's Information Criterion for Infinite Variance Autoregressive Processes},
     journal = {Ann. Statist.},
     volume = {17},
     number = {1},
     year = {1989},
     pages = { 824-840},
     language = {en},
     url = {http://dml.mathdoc.fr/item/1176347145}
}
Knight, Keith. Consistency of Akaike's Information Criterion for Infinite Variance Autoregressive Processes. Ann. Statist., Tome 17 (1989) no. 1, pp.  824-840. http://gdmltest.u-ga.fr/item/1176347145/