Scholarly record
MEASURING SOVEREIGN CREDIT RISK IN EU COUNTRIES USING AN ENSEMBLE OF CLASSIFIERS APPROACH
Abstract
In the age of globalization, international loans and sovereign default probabilities have become one of the main topics of finance. Many severe economic crises have occurred on financial markets in the last decades. Therefore, the problem of measuring countries default probabilities due to their external debt obligations has attracted the interest of many researchers. In recent literature there are many studies that use quantitative methods and machine learning techniques to predict countries defaults. Past studies on sovereign default prediction used many statistical and non-statistical approaches (machine learning), such as: logistic regression (two and more stages), discriminant analysis, support vector machines (SVM), neural networks, Bayesian models, classification trees, MARS (Multivariate Adaptive Regression Splines) and many more. In the recent years, in order to increase the classification efficiency of proposed models, the so-called ensemble classifiers are used. These techniques involve a combination of different methods and classifiers to significantly improve the effectiveness of models in assessing the prediction of bankruptcy or default on a loan. This paper presents the results of own research on the possibility of applying ensemble classifiers technique to international debt credit risk assessment for EU countries. The effectiveness of ensemble methods will be compared to some classic bankruptcy risk assessment methods. The ensemble model of sovereign default prediction was estimated in the 3-year forecast horizon and their efficiency was compared to certain individual models. The historical events of sovereign defaults come from Bank of Canada CRAG sovereign defaults database. The financial ratios used as predictors of internal debt defaults were estimated from the statistics of International Monetary Fund and World Bank databases. The estimated model will be used in sovereign default predictions for EU countries in year 2017.
Publication Impact Profile
Publication details
References17
Alp Ö. S., Büyükbebeci E., Cekic A. I., Özkurt F., Taylan P., Weber G.-W., CMARS and GAM & CQP - Modern optimization methods applied to international credit default prediction, Journal of Computational and Applied Mathematics, vol. 235, pp. 4639-4651, 2011.
Breiman L., Bagging predictors, Machine Learning, vol. 24, issue 2, pp. 123-140, 1996.
CEIC Indicators, [online]: https://www.ceicdata.com/en/indicators, CEIC Euromoney Institutional Investors Company [accessed: 15.01.2017].
Chakrabarti A., Zeaiter H., The determinants of sovereign default: A sensitivity analysis, International Review of Economics and Finance, vol. 33, pp. 300–318, 2014.
Ciarlone A., Trebeschi G., Designing an early warning system for debt crises, Emerging Markets Review, vol. 6, pp. 376–395, 2005.
Database of Sovereign Defaults, [online]: http://www.bankofcanada.ca/wp-content/ uploads/2015/05/crag-database-update-04-05-15.xlsx, Bank of Canada [accessed: 15.01.2017].
Dawood M., Horsewood N., Strobel F., Predicting sovereign debt crises: An Early Warning System approach, Journal of Financial Stability, vol. 28, pp. 16–28, 2017.
Debt and reserve related indicators of external vulnerability, [online]: http://www.imf.org/external/np/pdr/debtres/debtres.pdf, International Monetary Fund [accessed: 15.01.2017]
Fioramanti M., Predicting sovereign debt crises using artificial neural networks: A comparative approach, Journal of Financial Stability, vol. 4, pp. 149–164, 2008.
Freund Y., Schapire R. E., A short introduction to boosting, Journal of Japanese Society for Artificial Intelligence, vol. 14, issue 5, pp. 771-780, 1999.
Lee H.-H., Shih K., Wang K., Measuring sovereign credit risk using a structural model Approach, Review of Quantitative Finance and Accounting, vol. 47, pp. 1097– 1128, 2016.
Manasse P., Roubini N., Schimmelpfennig A., Predicting Sovereign Debt Crises, IMF Working Paper, No. 221, International Monetary Fund, pp. 1-41, 2003.
Saeys Y., Inza I., Larranaga P., A review of feature selection techniques in bioinformatics, Bioinformatics, vol. 23, issue 19, pp. 2507-2517, 2007.
Szetela B., Mentel G., Brożyna J., In search of insolvency among European countries, Economic Research-Ekonomska Istraživanja, vol. 29, pp. 839-856, 2016.
Wolpert D. H., Stacked generalization, Neural Networks, vol. 5, issue 2, pp. 241- 259, 1992.
World Development Indicators, [online]: http://databank.worldbank.org/data/ home.aspx, World Bank [accessed: 15.01.2017].
Zhou Z.-H., Ensemble methods. Foundations and algorithms, CRC Press, Boca Raton, 2012.
View or Download full articleAccess options
SWS access login
Login as SWS Scientific CommitteeLogin as SWS Scientific PartnerLogin as SWS AuthorAuthors and approved SWS contributors will read and export their own linked papers after identity matching by SWS profile, email and SGEM GlobalID.
For librarian assistance: [email protected]
Purchase Instant Access
- Article can be downloaded after successful payment.
- Article may be used according to SWS library access terms.
- Article cannot be redistributed.
