Weak identification in the ESTAR model and a new model

Research output: Contribution to journalArticleResearchpeer review

Authors

  • Florian Heinen
  • Stefanie Michael
  • Philipp Sibbertsen

Research Organisations

View graph of relations

Details

Original languageEnglish
Pages (from-to)238-261
Number of pages24
JournalJournal of time series analysis
Volume34
Issue number2
Early online date8 Nov 2012
Publication statusPublished - Mar 2013

Abstract

Determining good parameter estimates in (exponential smooth transition autoregressive) models is known to be difficult. We show that the phenomena of getting strongly biased estimators is a consequence of the so-called identification problem, the problem of properly distinguishing the transition function in relation to extreme parameter combinations. This happens in particular for either very small or very large values of the error term variance. Furthermore, we introduce a new alternative model - the TSTAR model - which has similar properties as the ESTAR model but reduces the effects of the identification problem. We also derive a linearity and a unit root test for this model.

Keywords

    Linearity testing, Nonlinearities, Real exchange rates, Smooth transition, Unit root testing

ASJC Scopus subject areas

Cite this

Weak identification in the ESTAR model and a new model. / Heinen, Florian; Michael, Stefanie; Sibbertsen, Philipp.
In: Journal of time series analysis, Vol. 34, No. 2, 03.2013, p. 238-261.

Research output: Contribution to journalArticleResearchpeer review

Heinen F, Michael S, Sibbertsen P. Weak identification in the ESTAR model and a new model. Journal of time series analysis. 2013 Mar;34(2):238-261. Epub 2012 Nov 8. doi: 10.1111/jtsa.12008
Heinen, Florian ; Michael, Stefanie ; Sibbertsen, Philipp. / Weak identification in the ESTAR model and a new model. In: Journal of time series analysis. 2013 ; Vol. 34, No. 2. pp. 238-261.
Download
@article{9267c3644b424579a892190e6df97879,
title = "Weak identification in the ESTAR model and a new model",
abstract = "Determining good parameter estimates in (exponential smooth transition autoregressive) models is known to be difficult. We show that the phenomena of getting strongly biased estimators is a consequence of the so-called identification problem, the problem of properly distinguishing the transition function in relation to extreme parameter combinations. This happens in particular for either very small or very large values of the error term variance. Furthermore, we introduce a new alternative model - the TSTAR model - which has similar properties as the ESTAR model but reduces the effects of the identification problem. We also derive a linearity and a unit root test for this model.",
keywords = "Linearity testing, Nonlinearities, Real exchange rates, Smooth transition, Unit root testing",
author = "Florian Heinen and Stefanie Michael and Philipp Sibbertsen",
year = "2013",
month = mar,
doi = "10.1111/jtsa.12008",
language = "English",
volume = "34",
pages = "238--261",
journal = "Journal of time series analysis",
issn = "0143-9782",
publisher = "Wiley-Blackwell",
number = "2",

}

Download

TY - JOUR

T1 - Weak identification in the ESTAR model and a new model

AU - Heinen, Florian

AU - Michael, Stefanie

AU - Sibbertsen, Philipp

PY - 2013/3

Y1 - 2013/3

N2 - Determining good parameter estimates in (exponential smooth transition autoregressive) models is known to be difficult. We show that the phenomena of getting strongly biased estimators is a consequence of the so-called identification problem, the problem of properly distinguishing the transition function in relation to extreme parameter combinations. This happens in particular for either very small or very large values of the error term variance. Furthermore, we introduce a new alternative model - the TSTAR model - which has similar properties as the ESTAR model but reduces the effects of the identification problem. We also derive a linearity and a unit root test for this model.

AB - Determining good parameter estimates in (exponential smooth transition autoregressive) models is known to be difficult. We show that the phenomena of getting strongly biased estimators is a consequence of the so-called identification problem, the problem of properly distinguishing the transition function in relation to extreme parameter combinations. This happens in particular for either very small or very large values of the error term variance. Furthermore, we introduce a new alternative model - the TSTAR model - which has similar properties as the ESTAR model but reduces the effects of the identification problem. We also derive a linearity and a unit root test for this model.

KW - Linearity testing

KW - Nonlinearities

KW - Real exchange rates

KW - Smooth transition

KW - Unit root testing

UR - http://www.scopus.com/inward/record.url?scp=84874193970&partnerID=8YFLogxK

U2 - 10.1111/jtsa.12008

DO - 10.1111/jtsa.12008

M3 - Article

AN - SCOPUS:84874193970

VL - 34

SP - 238

EP - 261

JO - Journal of time series analysis

JF - Journal of time series analysis

SN - 0143-9782

IS - 2

ER -