Περίληψη : | Probability of Default (PD) is a crucial credit risk parameter. International accords, like Basel II, have motivated banks and credit institutions to adopt objective systems of evaluating and monitoring the PD. The aim of this study is to give a review of the various statistical methodologies which have been used in PD estimation and apply a markov chain approach in real data. The methodologies discussed include linear regression, discriminant analysis, logistic regression, probit regression, survival analysis and stochastic processes. The application refers to retail unsecured loans of a major Greek bank.It focuses in the stochastic behavior of the financial states of the loans. It is supposed that a first order Markov chain model describes sufficiently the transitions from one state to another. It is inferred that the chain is non – homogeneous. Poisson regression models are estimated in order to calculate the limiting transition matrix, the limiting state probabilities and the Probability of Default.
|
---|