Linear and non-linear unit root testing in the presence of heavy-tailed GARCH: a finite-sample simulation analysis Online publication date: Wed, 17-Apr-2013
by Steve Cook
International Journal of Computational Economics and Econometrics (IJCEE), Vol. 2, No. 3/4, 2012
Abstract: Using numerical simulation, recent research on the properties of unit root tests in the presence of generalised autoregressive conditional heteroskedasticity (GARCH) is extended. The principal development concerns consideration of relative properties of linear and non-linear unit root tests in the presence of heavy-tailed GARCH innovations via the use of the student-t distribution and the generalised error distribution. The results obtained show the non-linear unit root test of Kapetanios et al. (2003) to suffer far greater finite-sample size distortion than the linear Dickey-Fuller test. The impact of heteroskedasticity consistent covariance matrix estimators is also considered. It is found that these 'robust' methods are unable to guarantee size correction in the presence of heavy-tailed GARCH processes and have differing effects depending upon the exact estimator used and whether they are applied to linear or non-linear unit root tests.
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Computational Economics and Econometrics (IJCEE):
Login with your Inderscience username and password:
Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.
If you still need assistance, please email subs@inderscience.com