This paper shows that the co-movement of public revenues in the European Monetary Union (EMU) is driven by an unobserved common factor. Our empirical analysis uses yearly data covering the period 1970–2014 for 12 selected EMU member countries. We have found that this common component has a significant impact on public revenues in the majority of the countries. We highlight this common pattern in a dynamic factor model (DFM). Since this factor is unobservable, it is difficult to agree on what it represents. We argue that the latent factor that emerges from the two different empirical approaches used might have a composite nature, being the result of both the more general convergence of the economic cycles of the countries in the area and the increasingly better tuned tax structure. However, the original aspect of our paper is the use of a back-propagation neural networks (BPNN)-DF model to test the results of the time-series. At the level of computer programming, the results obtained represent the first empirical demonstration of the latent factor’s presence.

Magazzino, C., Mele, M. (2021). A dynamic factor and neural networks analysis of the co-movement of public revenues in the EMU. ITALIAN ECONOMIC JOURNAL [10.1007/s40797-021-00155-2].

A dynamic factor and neural networks analysis of the co-movement of public revenues in the EMU

Magazzino, Cosimo;Mele, Marco
2021-01-01

Abstract

This paper shows that the co-movement of public revenues in the European Monetary Union (EMU) is driven by an unobserved common factor. Our empirical analysis uses yearly data covering the period 1970–2014 for 12 selected EMU member countries. We have found that this common component has a significant impact on public revenues in the majority of the countries. We highlight this common pattern in a dynamic factor model (DFM). Since this factor is unobservable, it is difficult to agree on what it represents. We argue that the latent factor that emerges from the two different empirical approaches used might have a composite nature, being the result of both the more general convergence of the economic cycles of the countries in the area and the increasingly better tuned tax structure. However, the original aspect of our paper is the use of a back-propagation neural networks (BPNN)-DF model to test the results of the time-series. At the level of computer programming, the results obtained represent the first empirical demonstration of the latent factor’s presence.
Magazzino, C., Mele, M. (2021). A dynamic factor and neural networks analysis of the co-movement of public revenues in the EMU. ITALIAN ECONOMIC JOURNAL [10.1007/s40797-021-00155-2].
File in questo prodotto:
File Dimensione Formato  
Magazzino-Mele2021_Article_ADynamicFactorAndNeuralNetwork.pdf

accesso aperto

Tipologia: Versione Editoriale (PDF)
Licenza: Creative commons
Dimensione 3.19 MB
Formato Adobe PDF
3.19 MB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11590/388391
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 0
  • ???jsp.display-item.citation.isi??? 6
social impact