An Artificial Intelligence Approach to Shadow Rating

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.JEL Classification codes: C45, C55, G24 AMS Classification codes: 62M45, 91G40

Machine Learning approach for Credit Scoring

“…Starting from these studies, we propose a sophisticated framework of machine learning models which, on the basis of company annual (end-of-year) financial statements coupled with relevant macroeconomic indicators, attempts to classify the status of a company (performing -"in-bonis" -or defaulted) and to build a robust rating system in which each rating class will be matched to an internally calibrated default probability. In this regard, here the target variable is different from a previous work by some of the authors [6], where the goal was to predict the credit rating that Moody's would assign, according to an approach commonly called "shadow rating". The novelty of our approach lies in the combination of data preprocessing algorithms, responsible for feature engineering and feature selection, and a core model architecture made of a concatenation of a Boosted Tree default classifier, a probability calibrator and a rating attribution system based on genetic algorithm.…”

“…a 3-layer encoder and 3-layer decoder6 The reconstruction loss is the loss function (usually either the mean-squared error or cross-entropy between the reconstructed output and the input) which penalizes the network for creating outputs different from the original input…”