Credit Risk Modeling: PD, LGD, EAD & Expected Credit Loss
Expected loss on a loan book, EAD with conversion factors, LGD from collateral and recoveries, migration matrices to cumulative PD, 12-month vs lifetime ECL with scenarios, IFRS 9 staging, a logistic scorecard with PDO scaling and portfolio coverage: 8 labs on one bank's book.
The first lesson is free — no account needed.
About this course
A credit risk modeling course that measures one bank's loan book in a spreadsheet, for credit analysts, risk and finance teams and students. You calculate expected loss as PD × LGD × EAD loan by loan, estimate a credit conversion factor for undrawn lines, derive LGD from collateral, haircuts, workout costs and the time value of the recovery, roll a borrower through a rating migration matrix to cumulative and marginal PD, compare the 12-month with the lifetime ECL with discounting and three weighted scenarios, assign loans to IFRS 9 stages with a PD test, the 30-days-past-due backstop and a watch list, turn a logistic score into a PD and into scorecard points with PDO scaling and report the coverage ratio by segment. The models are simplified and educational, not a bank's or regulator's methodology. The first lab is free.
Lessons
- Expected Loss = PD × LGD × EAD on a Loan BookFree
- EAD with a Credit Conversion Factor for Undrawn Lines Pro
- LGD from Collateral, Haircuts and Recovery Costs Pro
- Rating Migration Matrix to Cumulative PD Pro
- 12-Month vs Lifetime ECL with Discounting and Scenarios Pro
- IFRS 9 Staging: SICR Rules and Stage 1/2/3 ECL Pro
- Logistic Scorecard: Score → PD and Points to Double the Odds Pro
- Portfolio ECL and Coverage Ratio by Segment Pro