Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/89590 
Year of Publication: 
2013
Series/Report no.: 
Nota di Lavoro No. 105.2013
Publisher: 
Fondazione Eni Enrico Mattei (FEEM), Milano
Abstract: 
The present work applies several advanced spectral methods to the analysis of macroeconomic fluctuations in three countries of the European Union: Italy, The Netherlands, and the United Kingdom. We focus here in particular on singular-spectrum analysis (SSA), which provides valuable spatial and frequency information of multivariate data and that goes far beyond a pure analysis in the time domain. The spectral methods discussed here are well established in the geosciences and life sciences, but not yet widespread in quantitative economics. In particular, they enable one to identify and describe nonlinear trends and dominant cyclesincluding seasonal and interannual components that characterize the deterministic behavior of each time series. These tools have already proven their robustness in the application on short and noisy data, and we demonstrate their usefulness in the analysis of the macroeconomic indicators of these three countries. We explore several fundamental indicators of the countries' real aggregate economy in a univariate, as well as a multivariate setting. Starting with individual single-channel analysis, we are able to identify similar spectral components among the analyzed indicators. Next, we consider combinations of indicators and countries, in order to take different effects of comovements into account. Since business cycles are cross-national phenomena, which show common characteristics across countries, our aim is to uncover hidden global behavior across the European economies. Results are compared with previous findings on the U.S. indicators (Groth et al., 2012). Finally, the analysis is extended to include several indicators from the U.S. economy, in order to examine its influence on the European market.
Subjects: 
Advanced Spectral Methods
European Business Cycle
Frequency Domain
Time Domain
JEL: 
C15
C60
E32
Document Type: 
Working Paper

Files in This Item:
File
Size





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