An Artificial Intelligence Approach to Shadow Rating

18 Pages Posted: 10 Jan 2020

Date Written: December 20, 2019

Abstract

We analyse the effectiveness of modern deep learning techniques in predicting credit ratings over a universe of thousands of global corporate entities obligations when compared to most popular, traditional machine-learning approaches such as linear models and tree-based classifiers. Our results show a adequate accuracy over different rating classes when applying categorical embeddings to artificial neural networks (ANN) architectures.

Keywords: Rating Model, Shadow Rating, Artificial Intelligence, Machine Learning, Explainable AI

JEL Classification: C45, C55, G24

Suggested Citation

Provenzano, Angela Rita and Trifirò, Daniele and Jean, Nicola and Le Pera, Giacomo and Spadaccino, Maurizio and Massaron, Luca and Nordio, Claudio, An Artificial Intelligence Approach to Shadow Rating (December 20, 2019). Available at SSRN: https://ssrn.com/abstract=3507420 or http://dx.doi.org/10.2139/ssrn.3507420

Angela Rita Provenzano (Contact Author)

illimity bank ( email )

Via Soperga
Milano
Italy

Daniele Trifirò

illimity bank ( email )

Via Soperga
Milano
Italy

Nicola Jean

illimity bank

Via Soperga
Milano
Italy

Giacomo Le Pera

illimity Bank ( email )

Via Soperga
Milano
Italy

Maurizio Spadaccino

illimity bank ( email )

Via Soperga
Milano
Italy

Luca Massaron

illimity bank ( email )

Via Soperga
Milano
Italy

Claudio Nordio

illimity bank ( email )

Milano
Italy

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
90
Abstract Views
359
rank
334,737
PlumX Metrics