National and Kapodistrian University of Athens
Greek crisis and credit risk: predictive ability of the credit risk models and stress testing
Abstract
dc:descriptionThis thesis studies corporate credit risk in the Greek equity market by treating financial distress as a dynamic process rather than as a rare binary default event. The Greek economy constitutes a particularly informative case, as the sovereign debt crisis, the banking crisis, the prolonged recession, and the COVID-19 shock generated repeated regime breaks and persistent financial fragility among firms. In this environment, formal bankruptcies and delistings capture only part of underlying economic distress, while institutional features often delay or distort observable default outcomes. Therefore, the analysis shifts the focus from static default prediction to the evolution of firms’ financial condition over time. Altman’s Z-Score is used as the central organizing metric, not only through its traditional zone classification but primarily as a continuous indicator whose changes, direction of movement and future levels convey information about evolving credit quality. The empirical framework combines classical credit risk models, panel econometric techniques, and machine learning methods applied to a long panel of non-financial firms listed on the Athens Stock Exchange from the mid-2000s to the early 2020s. The results highlight the dominant role of profitability and liquidity in driving distress dynamics, strong persistence and mean reversion in credit quality, and the importance of macroeconomic regimes in shaping firm-level outcomes. Predictive models show that the direction of financial improvement or deterioration is substantially more predictable than the magnitude of changes, while machine learning methods improve out-of-sample performance but remain sensitive to extreme outcomes. Overall, the thesis contributes a unified framework for credit risk monitoring in small, crisis-prone economies, emphasizing the need to complement static default indicators with dynamic, interpretable measures of financial distress, and explicitly acknowledging the limits of prediction under structural change.
Degree
thesis:*- Grantor dc:publisher
- National and Kapodistrian University of Athens
- Year dc:date
- 2026
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Tsoutsou, Alexandra
- Τσούτσου, Αλεξάνδρα
Subjects
dc:subject × 18- Πιστωτικός κίνδυνος επιχειρήσεων
- Οικονομική δυσχέρεια
- Δείκτης Altman Z-Score
- Δεδομένα πάνελ
- Μηχανική μάθηση
- Πρόβλεψη αθέτησης υποχρεώσεων
- Corporate credit risk
- Financial distress
- Altman Z-score
- Panel data
- Machine learning
- Default prediction
- Οικονομικά και Επιχειρήσεις
- Κοινωνικές Επιστήμες
- Χρηματοοικονομικά
- Economics and Business
- Social Sciences
- Finance
Rights
- Language dc:language
- eng
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/62491