Can One Really Estimate Nonstationary GARCH Models ?
Jensen and Rahbek (2004a) claim that consistency and asymptotic normality hold for the quasi-maximumlikelihood estimator (QMLE) of (!0, 0) in nonstationary ARCH(1) models. In fact their result onlyconcerns a constrained QMLE, in which the intercept is fixed, and under a reinforced nonstationaritycondition. Under this condition, we prove that the standard QMLE of 0 is strongly consistent andasymptotically normal. Numerical experiments reveal that QMLE of !0 is likely to be inconsistent.
|Date of creation:||2008|
|Date of revision:|
|Contact details of provider:|| Postal: 15 Boulevard Gabriel Peri 92245 Malakoff Cedex|
Phone: 01 41 17 60 81
Web page: http://www.crest.fr
More information through EDIRC
When requesting a correction, please mention this item's handle: RePEc:crs:wpaper:2008-06. See general information about how to correct material in RePEc.
For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: (Florian Sallaberry)
If references are entirely missing, you can add them using this form.