Predicting sovereign debt crises: an early warning system approach

Mary Dawood, Nicholas Horsewood, Frank Strobel*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

21 Citations (Scopus)
530 Downloads (Pure)

Abstract

In light of the renewed challenge to construct effective "Early Warning Systems" for sovereign debt crises, we empirically evaluate the predictive power of econometric models developed so far across developed and emerging country regions. We propose a different specification of the crisis variable that allows for the prediction of new crisis onsets as well as duration, and develop a more powerful dynamic-recursive forecasting technique to generate more accurate out-of-sample warning signals of sovereign debt crises. Our results are shown to be more accurate compared to the ones found in the existing literature.
Original languageEnglish
Pages (from-to)16-28
Number of pages13
JournalJournal of Financial Stability
Volume28
Early online date27 Nov 2016
DOIs
Publication statusPublished - Feb 2017

Keywords

  • Sovereign debt crisis
  • Early Warning System
  • Logit
  • Dynamic signal extraction
  • Dynamic-recursive forecasting

Fingerprint

Dive into the research topics of 'Predicting sovereign debt crises: an early warning system approach'. Together they form a unique fingerprint.

Cite this