Front Matter Page
Fiscal Affairs Department
Contents
I. Introduction
II. Does Debt Matter? Lessons from the Literature
III. Data
A. Measuring Fiscal Crises
B. Predictors
IV. Methodology
A. Variable selection
B. Assessing variable importance
C. Studying interactions and nonlinearities
V. Results
A. Variable selection
B. Variable importance
C. An analysis of selected predictors
VI. Conclusion
References
Tables
Table 1. Fiscal Crises Episodes (1980–2016)
Table 2. Out-of-Sample Performance of Alternative Feature Selection Algorithms
Figures
Figure 1. Predictors of Fiscal Crises in the Literature
Figure 2. Most Common Predictors in the Literature
Figure 3. Countries with Fiscal Crises, 1980–2016
Figure 4. Overlap with Other Crises, 1980–2016
Figure 5. Debt Statistics: Country Coverage, 1980–2016
Figure 6. Public and Public External Debt, 1980–2016
Figure 7. Interest-Growth Differential, 1980–2016
Figure 8. Feature Selection Algorithms
Figure 9. Robustness in Variable Selection
Figure 10. Variable Importance by Group of Predictors
Figure 11. Contribution to Probability of a Crisis
Figure 12. Partial Dependence Plots1/ and Event Studies
Figure 13. Overall Interaction Strength
Figure 14 Top-10 Interactions with Public
Figure 15. Bivariate Partial Dependent Plots: Public External Debt and r-g
Figure 16. Inflation: Univariate Partial Depedence Plots
Figure 17. Inflation and Public External Debt: Bivariate Partial Dependence Plots
Figure 18. Current Account: Univariate Partial Dependence Plots
Figure 19. Current Account and Public External Debt: Bivariate Partial Dependence Plots
Figure 20. Credit Gap: Univariate Partial Dependence Plots
Figure 21. Credit Gap and Public External Debt: Bivariate Partial Dependence Plots
Appendixes
1. Literature Review
2. Fiscal Crisis: Definitions and Data Sources
3. Sample of Countries
4. Data: Definition, Sources, and Predictor Groupings
5. Methodological Details
Appendix Table 5.1. Out-of-Sample Performance