Localized Level Crossing Random Walk Test Robust to the Presence of Structural Breaks

Vitali Alexeev and Alex Maynard

No 1001, Working Papers from University of Guelph, Department of Economics and Finance

Abstract: We propose a modified version of the nonparametric level crossing random walk test, in which the crossing level is determined locally. This modification results in a test that is robust to unknown multiple structural breaks in the level and slope of the trend function under both the null and alternative hypothesis. No knowledge regarding the number or timing of the breaks is required. An algorithm is proposed to select the degree of localization in order to maximize bootstrapped power in a proximate model. A computational procedure is then developed to adjust the critical values for the effect of this selection procedure by replicating it under the null hypothesis. The test is applied to Canadian nominal inflation and nominal interest rate series with implications for the Fisher hypothesis.

Keywords: Level crossing; random walk; structural breaks; unit root; robustness (search for similar items in EconPapers)
JEL-codes: C12 C14 C22 (search for similar items in EconPapers)
Pages: 37 pages
Date: 2010
New Economics Papers: this item is included in nep-ecm and nep-ets
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
http://www.uoguelph.ca/economics/repec/workingpapers/2010/2010-01.pdf (application/pdf)

Related works:
Journal Article: Localized level crossing random walk test robust to the presence of structural breaks (2012)
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:gue:guelph:2010-01.

Access Statistics for this paper

More papers in Working Papers from University of Guelph, Department of Economics and Finance Contact information at EDIRC.
Bibliographic data for series maintained by Stephen Kosempel ().