Time matters: how default resolution times impact final loss rates
Using access to a unique bank loss data base, we show positive dependencies of default resolution times (DRTs) of defaulted bank loan contracts and final loan loss rates (losses given default, LGDs). Due to this interconnection, LGD predictions made at the time of default and during resolution are subject to censoring. Pure (standard) LGD models are not able to capture effects of censoring and entail parameter distortions and, thus, an underestimation of average LGDs. In this paper, we develop a Bayesian hierarchical modeling framework for DRTs and LGDs which enables adequate unconditional LGD predictions and consistent LGD predictions conditional on the time in default in accordance with recent regulatory guidelines within one modeling framework. The proposed method is applicable to duration processes in general where the final outcomes depend on the duration of the process and are affected by censoring. By this means, potential parameter inconsistencies are overcome to ensure adequate predictions.
Download the PDF version of this document
- Published on 06/05/2019
- PDF (312.77 KB)
Updated on: 06/05/2019 17:03