Predicting Sovereign Debt Crises with Fuzzy Decision Trees

Abstract

Considering the great capacity of data mining techniques to extract useful information from large databases and to manage heterogeneous variables, this paper uses Fuzzy C4.5 Decision Trees for the prediction of sovereign debt crises. To this end, prediction models have been constructed for different regions, and another global model for the whole world. The results obtained show that Fuzzy C4.5 Decision Trees method overcomes the predictive power of existing models in the previous literature and provides more explanatory information on the reasons that cause sovereign debt crises.