On one-sided score tests : an application to tests of GARCH effects
Files
(Published version)
Date
2013
Authors
Lu, Z.H.
Editors
Advisors
Journal Title
Journal ISSN
Volume Title
Type:
Working paper
Citation
Statement of Responsibility
Conference Name
Abstract
Score tests are widely used inference tools in econometrics and statistics because of their advantage of requiring only restricted estimates, which may be much easier to obtain than otherwise. In this paper we first study existing one-sided score based tests. Some properties are presented. We also show that there are some issues that may affect the validity of existing score based tests. Second, we propose one-sided score tests under a sufficient condition. The tests are shown to be locally asymptotically uniformly more powerful than usual two-sided score tests. The sufficient condition is shown to be guaranteed in the case of testing GARCH effects. The advantages of our tests over existing tests are revealed in our simulation studies on tests of GARCH effects. Extensions of our tests in the absence of the sufficient condition are discussed.
School/Discipline
Dissertation Note
Provenance
Description
Access Status
Rights
Copyright [2013] Zeng-Hua Lu. Not for distribution without permission of author.