In the LGD Model flow presented in the figure 4.13 in the book "Developing Credit Risk Models Using SAS Enterprise Miner and SAS/STAT: Theory and Application" which is partially available on the web:


The authors suggest to develop multiple scores (based on logistic and linear regression) depending on the account characteristics (i.e. if it has caused a loss or if it defaulted). By the end of the process, the whole development data is aggregated in order to be modeled and obtain the 'final' score. I don't understand the value of doing all this work. Why not just model the default accounts and suppose that non default accounts will act likewise? In addition, by the end of this process the author mentions that "The total number of payers from the augmented data are filtered. Another linear regression model is applied on the filtered data to predict the amount of loss. While scoring, only the non-defaulted accounts are scored." Do I have to understand that the objectif of this LGD score is to predict the amount of loss of non default accounts?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.