Loughborough University
Leicestershire, UK
LE11 3TU
+44 (0)1509 263171
Loughborough University

Loughborough University Institutional Repository

Please use this identifier to cite or link to this item: https://dspace.lboro.ac.uk/2134/2056

Title: Asymmetric adjustment and bias in estimation of an equilibrium relationship from a cointegrating regression
Authors: Holly, Sean
Turner, Paul
Weeks, Melvyn
Issue Date: 2003
Publisher: © Kluwer (Springer)
Citation: HOLLY, S.,TURNER, P.and WEEKS, M., 2003. Asymmetric adjustment and bias in estimation of an equilibrium relationship from a cointegrating regression. Computational Economics, 21, pp.195-202.
Abstract: This paper uses Monte Carlo methods to investigate the effects of asymmetric adjustment on estimates of the parameters of the equilibrium relationship between a set of variables.We demonstrate that simple least squares estimates and the implicit estimates from a symmetric error correction model both lead to biases in the constant term. This bias increases with the size of the asymmetry and shows no tendency to decline with the sample size. We also show that if the biased estimates of the equilibrium relationship are then used to devide the sample into different regimes to test for assymmetric adjustment, then the resulting test has low power. The power of tests for asymmetry can be increased significantly by using simultaneous estimation of the parameters of the equilibrium relationship and the asymmetric adjustment process.
Description: This is Restricted Access. The article was published in the journal, Computational Economics [© Springer] and is available at: http://www.springerlink.com/openurl.asp?genre=journal&issn=0927-7099
URI: https://dspace.lboro.ac.uk/2134/2056
ISSN: 0927-7099
Appears in Collections:Closed Access (Economics)

Files associated with this item:

File Description SizeFormat
Computational_Economics_2003.pdfArticle71.92 kBAdobe PDFView/Open

 

SFX Query

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.