Abstract
In this article, multivariate statistical models are derived for forecasting the emergence of arrears of debt-service among less-developed countries. The emphasis of the article is on prediction, and the criterion variable differs from that generally used in earlier research, which focuses on rescheduling. Out-of-sample tests are used to establish the ex ante predictive utility of the derived models, which are found to have low error rates. Moreover, the predictive ability appears to be stable over time. © 1994 Kluwer Academic Publishers.
Original language | English |
---|---|
Pages (from-to) | 339-356 |
Number of pages | 17 |
Journal | Review of Quantitative Finance and Accounting |
Volume | 4 |
Issue number | 4 |
Publication status | Published - Dec 1994 |
Keywords
- arrears
- country risk
- discriminant
- logit
- prediction
- rescheduling