Structured Dictionary Learning of Rating Migration Matrices for Credit Risk Modeling

Michaël Allouche (1) , Emmanuel Gobet (1) , Clara Lage (1) , Edwin Mangin (2)

1 CMAP - Centre de Mathématiques Appliquées - Ecole Polytechnique
2 BNPP

Rating Migration Matrix is a crux to assess credit risks. Modeling and predicting these matrices are then an issue of great importance for risk managers in any financial institution. As a challenger to usual parametric modeling approaches, we propose a new structured dictionary learning model with auto-regressive regularization that is able to meet key expectations and constraints: small amount of data, fast evolution in time of these matrices, economic interpretability of the calibrated model. To show the model applicability, we present a numerical test with real data. The source code and the data are available at https://github.com/michael-allouche/ dictionary-learning-RMM.git for the sake of reproducibility of our research.

Mots clés: Rating Migration Matrix; Dictionary learning; auto-regressive modeling; interpretability

Domaines: Machine Learning [stat.ML]; Finance quantitative [q-fin.CP]

Michaël Allouche, Emmanuel Gobet, Clara Lage, Edwin Mangin. Structured Dictionary Learning of Rating Migration Matrices for Credit Risk Modeling. Computational Statistics, 2024, ⟨10.1007/s00180-023-01449-y⟩. ⟨hal-03715954v2⟩