Advances and Applications in Statistics
Volume 13, Issue 2, Pages 205 - 223
(December 2009)
|
|
TESTING OF I(d) STATISTICAL MODELS WITH NON-PARAMETRIC DISTURBANCES
Luis A. Gil-Alana (Spain)
|
Abstract: In this article, we propose the use of a modified version of a procedure of Robinson [31] for testing I(d) statistical models. However, instead of assuming a purely parametric model for the I(0) disturbances, as it is the case in Robinson [31], we use a non-parametric approach, based on an estimate of its long run variance. Finite-sample critical values of the new tests are computed, and several Monte Carlo experiments conducted across the paper show that the tests perform relatively well in finite samples. An empirical application, based on an extended version of Nelson and Plosser’s [24] dataset, is also carried out at the end of the article. |
Keywords and phrases: fractional integration, long memory. |
|
Number of Downloads: 379 | Number of Views: 1393 |
|