Corporate Default Prediction Model Averaging: A Normative Linear Pooling Approach

Focusing on credit risk modelling, this paper introduces a novel approach for ensemble modelling based on a nor-mative linear pooling. Models are first classified as dominant and competitive, and the pooling is run using thecompetitive models only. Numerical experiments based on parametric (logit, Bayesian model averaging) and non-parametric (classification tree, random forest, bagging, boosting) model comparison shows that the proposed en-semble performs better than alternative approaches, in particular when different modelling cultures are mixedtogether (logit and classification tree).