Publicação
Estimation of probability of default for low default portfolios
| Resumo: | Under Basel II, banks can use internal models to calculate their regulatory capital for credit risk, being the probability of default (PD) one of the fundamental parameters used in the quantification of credit risk. The Basel Committee on Banking Supervision requires financial institutions to incorporate a margin of conservatism to their estimates for PD in specific cases, namely in low default portfolios (LDPs). As an attempt to overcome the problems with the PD estimates on LDPs, some authors proposed different approaches, standing out the upper confidence based and the Bayesian approaches. The upper confidence based approach is based on the most prudent estimation principle, while the Bayesian approach specifies a prior distribution function over parameters that must be estimated. |
|---|---|
| Autores principais: | Carriço, Maria Rita Moura Varela |
| Assunto: | Basel II Bayesian estimation credit risk low default portfolio probability of default upper confidence bound Acordo de Basileia II estimação Bayesiana risco de crédito carteiras de baixa sinistralidade probabilidade de incumprimento limite de confiança superior |
| Ano: | 2022 |
| País: | Portugal |
| Tipo de documento: | dissertação de mestrado |
| Tipo de acesso: | acesso aberto |
| Instituição associada: | Universidade de Lisboa |
| Idioma: | inglês |
| Origem: | Repositório da Universidade de Lisboa |
| Resumo: | Under Basel II, banks can use internal models to calculate their regulatory capital for credit risk, being the probability of default (PD) one of the fundamental parameters used in the quantification of credit risk. The Basel Committee on Banking Supervision requires financial institutions to incorporate a margin of conservatism to their estimates for PD in specific cases, namely in low default portfolios (LDPs). As an attempt to overcome the problems with the PD estimates on LDPs, some authors proposed different approaches, standing out the upper confidence based and the Bayesian approaches. The upper confidence based approach is based on the most prudent estimation principle, while the Bayesian approach specifies a prior distribution function over parameters that must be estimated. |
|---|