Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/82441 
Year of Publication: 
2003
Series/Report no.: 
Sveriges Riksbank Working Paper Series No. 150
Publisher: 
Sveriges Riksbank, Stockholm
Abstract: 
A neglected aspect of the otherwise fairly well developed Bayesian analysis of cointegration is the point estimation of the cointegration space. It is pointed out here that, due to the well known non-identification of the cointegration vectors, the parameter space is not an inner product space and conventional Bayes estimators therefore stand without their usual decision theoretic foundation. We present a Bayes estimator of the cointegration space which takes the curved geometry of the parameter space into account. Contrary to many of the Bayes estimators used in the literature, this estimator is invariant to the ordering of the time series. A dimension invariant overall measure of cointegration space uncertainty is also proposed. A small simulation study shows that the Bayes estimator compares favorably to the maximum likelihood estimator.
Subjects: 
Bayesian inference
Cointegration analysis
Estimation
Grassman manifold
Subspaces.
JEL: 
C11
C13
C32
Document Type: 
Working Paper

Files in This Item:
File
Size
497.03 kB





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