We investigate the puzzle in the literature that various parametric loss given default (LGD) statistical models perform similarly by comparing their performance in a simulation framework. We find that, even using the full set of explanatory variables from the assumed data generating process,...

We apply multiple machine learning (ML) methods to model loss given default (LGD) for corporate debt using a common dataset that is cross-sectional but collected over different time periods and shows much variation over time. We investigate the efficacy of three cross-validation (CV) schemes for...

We use Markov chain Monte Carlo (MCMC) sampling to draw model coefficients to generate LGD distributions. We find that applying this Bayesian method on a sophisticated model that accounts for the bi-modal distribution of the LGDs, such as the zero-one-inflated beta (ZOIB) model, can generate LGD...

Background: We examine the signaling effect of borrowers’ social media behavior, especially self-disclosure behavior, on the default probability of money borrowers on a peer-to-peer (P2P) lending site. Method: We use a unique dataset that combines loan data from a large P2P lending site with...