Banks are exposed to different sources of risk which may affect its equity levels and profitability. The main risk approached is called ‘credit risk’, defined by the probability of a debtor to default with the agreed contractual terms. This risk allows us to quantify the expected loss regarding an individual credit, which depends on three different parameters: the probability of default (PD), the exposure at default (EAD) and the loss given default (LGD). The following study applies four econometric techniques to estimate the probability of default, namely, linear probability model, logit, probit and multiple discriminant analysis. The sample collected covers 200 mortgage residential loans, originated between 2000 and 2010, including contracts affected by a default event during 2011 and contracts free of default event during the same year.
Date of Award | 16 Mar 2015 |
---|
Original language | Portuguese |
---|
Awarding Institution | - Universidade Católica Portuguesa
|
---|
Supervisor | Ricardo Cruz (Supervisor) & Vitor Gomes (Co-Supervisor) |
---|
- Credit risk
- Probability of default
- Residential mortgage loans
- Mortgage default
- Loan-to-Value ratio (LTV)
- Loan-to-Income ratio (LTI)
- Multiple discriminant analysis
- Logistic regression
- Mestrado em Banca e Seguros
Probabilidade de default em crédito à habitação: aplicação de técnicas de estimação alternativas
Alegre, M. T. (Student). 16 Mar 2015
Student thesis: Master's Thesis