Bastos, João A., "Ensemble Predictions of Recovery Rates", Journal of Financial Services Research, (forthcoming).

Abstract: In many domains, the combined opinion of a committee of experts provides better decisions than the judgment of a single expert. This paper shows how to implement a successful ensemble strategy for predicting recovery rates on defaulted debts. Using data from Moody's Ultimate Recovery Database, it is shown that committees of models derived from the same regression method present better forecasts of recovery rates than a single model. More accurate predictions are observed whether we forecast bond or loan recoveries, and across the entire range of actual recovery values.

JEL Classification: G17, G21.

Keywords: Recovery rate, Loss given default, Forecasting, Ensemble learning, Credit risk.

Books Referenced in this paper:  (what is this?)

Download paper (619K PDF) 26 pages

Most Cited Books within Recoveries/LGD Papers

[