Please use this identifier to cite or link to this item: http://hdl.handle.net/10419/86056
Authors: 
Luginbuhl, Rob
Koopman, Siem Jan
Year of Publication: 
2003
Series/Report no.: 
Tinbergen Institute Discussion Paper 03-031/4
Abstract: 
Convergence in gross domestic product series of five European countriesis empirically identified using multivariate time series models that arebased on unobserved components with dynamic converging properties.We define convergence in terms of a decrease in dispersion over timeand model this decrease via mechanisms that allow for gradualreductions in the ranks of covariance matrices associated with thedisturbance vectors driving the unobserved components of the model.The inclusion of such convergence mechanisms within the formulation ofunobserved components makes the identification of various types ofconvergence possible.The common converging component model isestimated for the per capita gross domestic product of five Europeancountries: Germany, France, Italy, Spain and the Netherlands. It is foundthat convergence features in trends and cycles are present and areassociated with some key events in the history of European integration.
Subjects: 
Common trends and cycles
dynamic factor model
economic convergence
Kalman filter
multivariate unobserved components time series models
JEL: 
C13
C32
E32
Document Type: 
Working Paper

Files in This Item:
File
Size
435.41 kB





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