An Application of Clustering Analysis to International Private Indebtedness

Monteiro Andre, Carneiro Dionisio and Carlos Eduardo Pedreira ()
Additional contact information
Monteiro Andre: Gavea Investimentos
Carneiro Dionisio: PUC- RIO

Computational Economics from EconWPA

Abstract: This paper presents a procedure for clustering analysis that combines Kohone’s Self organizing Feature Map (SOFM) and statistical schemes. The idea is to cluster the data in two stages: run SOFM and then minimize the segmentation dispersion. The advantages of proposed procedure will be illustrated through a synthetic experiment and a real macroeconomic problem. The procedure is then used to explore the relationship between private indebtedness and some macroeconomic variables commonly used to measure macroeconomic performance. The experiences of thirty-nine countries in the early nineties are analyzed. The procedure outperformed others clustering techniques in the job of identifying consistent groups of countries from the economic and statistical viewpoints. It found out similarities in different countries concerning their respective levels of private indebtedness when added to well accepted parameters to measure macroeconomic performance.

Keywords: Vector quantization; Clustering; Self-Organizing Feature Map; Macroeconomic Performance; Private Indebtedness. (search for similar items in EconPapers)
JEL-codes: C8 (search for similar items in EconPapers)
Date: Written
Note: Type of Document - pdf; pages: 13. published at 'LEARNING AND NONLINEAR MODELS' ISSN 1676-2789 Vol. 1, No. 4, pp. 264-277, Dec 2004
View list of references

Downloads: (external link)
http://129.3.20.41/eps/comp/papers/0505/0505001.pdf (application/pdf)

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: http://EconPapers.repec.org/RePEc:wpa:wuwpco:0505001

Access Statistics for this paper

More papers in Computational Economics from EconWPA
Series data maintained by EconWPA ().