Survival analysis and its application on a retail portfolio

The bank regulation embodied in the Basel II Accord has opened-up a new era in estimating RR or complementary measure LGD in retail lending credit evaluation process. The flexibility of a bank in determining LGD tailored to the bank's specific portfolio is a good motivation to move from the Foundation to IRB approach which, on the other hand, imposes on the bank the need to know as much of LGD as possible. The thesis investigates the properties of SA models applied to RR in order to predict LGD for retail lending. In contrary to the standard regression techniques, the SA methods enable to work with cases where only partial information is available, utilizing thus all available information contained in gathered data. The work finally compares the results of the standard techniques such as LR and LOR, respectively, with SA techniques and discusses the pros and cons of the respective methods. The study is performed on a real dataset of a major Czech bank.