Predicting Sovereign Debt Crises : Literature Review

Abstract

The frequency and the impact power of sovereign debt crises create a need to develop a model which could be used to predict such events. The literature has developed two types of model: parametric and non-parametric. Their common name is the early warning systems. Despite the relative popularity of this subject recent crises suggest that none of the developed models is sufficiently effective, or its use not widespread enough to avoid further crises. This article provides an overview of the achievements in the field of early warning systems and tries to answer the question which of the models are most effective and which variables are most commonly used in the models.