Credit Risk Value and Expected Deficit Applying Copulas
This paper presents an application of Copula Theory to an Ecuadorian consumer credit portfolio. To be applied, first, the marginal distributions of the default rate and the amount of exposure were estimated based on historical information; then copulas were built, and Sklar’s Theorem was applied thr...
| Autor: | |
|---|---|
| Tipo de recurso: | artículo |
| Estado: | Versión publicada |
| Fecha de publicación: | 2021 |
| País: | Ecuador |
| Institución: | Universidad Andina Simón Bolivar |
| Repositorio: | Revista Estudios de la Gestión |
| Idioma: | español |
| OAI Identifier: | oai:revistas.uasb.edu.ec:article/2579 |
| Acceso en línea: | https://revistas.uasb.edu.ec/index.php/eg/article/view/2579 |
| Access Level: | acceso abierto |
| Palabra clave: | cópula riesgo de crédito valor en riesgo de crédito déficit esperado Copula credit risk value at credit risk expected deficit Cópula risco de crédito valor em risco de crédito |
| Sumario: | This paper presents an application of Copula Theory to an Ecuadorian consumer credit portfolio. To be applied, first, the marginal distributions of the default rate and the amount of exposure were estimated based on historical information; then copulas were built, and Sklar’s Theorem was applied through Models of Multivariate Distribution of Copulas (MVDC). Subsequently, by knowing the dependency structure, the total loss of the portfolio, maximum loss, Credit VaR and Expected Shortfall (ES) were estimated. Considering a confidence level of 99,5 % in normal market conditions in a month, the maximum loss that the portfolio can present is USD 18.65 million (Credit VaR). If any factor changes and market conditions worsen, once the maximum loss is exceeded, the expected loss after Credit VaR, that is, ES can reach a value of USD 21.49 million (15,22 % more than Credit VaR) . Finally, when comparing the estimates of the MVDC with the methodology of the Ecuadorian control body, it was shown that it underestimates the expected loss, risk indicators and extreme loss events. The failure to predict extreme events underestimates potential losses and increases risk levels. |
|---|