Arbeitspapier
Exploring the sources of loan default clustering using survival analysis with frailty
This paper investigates whether three microeconomic loan characteristics are sources of loan default clustering in the Mexican banking sector by employing survival analysis with frailty. Using a large sample of bank loan level data granted to micro, small and medium sized firms from January 2010 to 2018, we test whether classifying loans by the bank's systemic importance, industry or at individual firm level enhances the predictions of loans defaults. Our results show that loans granted by Domestic Systemically Important Banks contribute to the default clustering in micro and small firm loans. This is due to aggregate default rate levels and clusters that are large for these firms loans compared with loans provided to medium-sized firms. These findings have important implications for bank's expected loss management related to the correlated loan default risk.
- Sprache
-
Englisch
- Erschienen in
-
Series: Working Papers ; No. 2021-14
- Klassifikation
-
Wirtschaft
Forecasting Models; Simulation Methods
Duration Analysis; Optimal Timing Strategies
Single Equation Models; Single Variables: Discrete Regression and Qualitative Choice Models; Discrete Regressors; Proportions; Probabilities
Corporate Finance and Governance: Government Policy and Regulation
- Thema
-
Credit risk
Parametric survival analysis
Accelerated Failure Time (AFT) models
Shared frailty models
IFRS 9
- Ereignis
-
Geistige Schöpfung
- (wer)
-
Bátiz-Zuk, Enrique
Mohamed, Abdulkadir
Sánchez-Cajal, Fátima
- Ereignis
-
Veröffentlichung
- (wer)
-
Banco de México
- (wo)
-
Ciudad de México
- (wann)
-
2021
- Handle
- Letzte Aktualisierung
-
10.03.2025, 11:42 MEZ
Datenpartner
ZBW - Deutsche Zentralbibliothek für Wirtschaftswissenschaften - Leibniz-Informationszentrum Wirtschaft. Bei Fragen zum Objekt wenden Sie sich bitte an den Datenpartner.
Objekttyp
- Arbeitspapier
Beteiligte
- Bátiz-Zuk, Enrique
- Mohamed, Abdulkadir
- Sánchez-Cajal, Fátima
- Banco de México
Entstanden
- 2021