The authors propose a new method to design credit risk rating models for corporate entities using a meta-algorithm which exploits information embedded in expert-assigned credit ratings to rank customers.
The authors propose a stressed version of distance to default to measure time-varying corporate default risk in the event of a systematic stress scenario.
Stressing of migration matrixes for International Financial Reporting Standard 9 and Internal Capital Adequacy Assessment Process calculations
This paper demonstrates that correlation estimates are sensitive to model assumptions and estimation methodology by comparing three methods used to stress rating transition matrixes.
The authors demonstrate a nonlinear impact of loan and borrower characteristics when applying a GAM framework to personal loans taken from a Korean bank.