Tests de linéarité, spécification et estimation de modèles à seuil: une analyse comparée des méthodes de Tsay et de Hansen

Melika Ben Salem and Corinne Perraudin

Economie & Prévision, 2001, vol. 148, issue 2, 157-176

Abstract: This paper empirically studies threshold models and their application to time series. Threshold models are distinct in that they can generate different dynamics depending on the state of the economy. The non-linearity that they induce precludes the use of standard estimation methods. Linearity tests as opposed to the alternative threshold models in particular depend on unidentifiable nuisance parameters under the null hypothesis of linearity. The paper presents two different specification methods for threshold models and compares them based on size and power exercises. It then presents an application to GDP growth rates for eight OECD countries.

Date: 2001
References: Add references at CitEc
Citations: View citations in EconPapers (10)

Downloads: (external link)
http://www.cairn.info/load_pdf.php?ID_ARTICLE=ECOP_148_0157 (application/pdf)
http://www.cairn.info/revue-economie-et-prevision-1-2001-2-page-157.htm (text/html)
free

Related works:
Journal Article: Tests de linéarité, spécification et estimation de modèles à seuil: une analyse comparée des méthodes de Tsay et de Hansen (2001)
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:cai:ecoldc:ecop_148_0157

Access Statistics for this article

More articles in Economie & Prévision from La Documentation Française
Bibliographic data for series maintained by Jean-Baptiste de Vathaire ().