Subscribe to DSC Newsletter

Hi everyone,


Currently, I'm working on LGD calculation for a bank portfolio and to do this, I'm planning to build a recovery scorecard. Here I need to calculate the estimated recovery rates for different customers at the time of write off. I'm going with the OLS regression instead of logistic regression because of some reasons. But the recovery rate distribution is a beta distribution. Can someone please help on how I can transform this to Normal distribution so that I can run OLS regression.


Also, any suggestions on other ways to calculate the LGD ?



Views: 344

Reply to This

Replies to This Discussion

You may attempt with Box Cox transformation. vide Wikipedia.


On Data Science Central

© 2020 is a subsidiary and dedicated channel of Data Science Central LLC   Powered by

Badges  |  Report an Issue  |  Privacy Policy  |  Terms of Service