Deep Learning Credit Risk Modeling

Forthcoming Journal of Fixed Income

Posted: 4 Aug 2020 Last revised: 11 Jun 2021

See all articles by Gerardo Manzo

Gerardo Manzo


Xiao Qiao

School of Data Science, City University of Hong Kong; Paraconic Technologies US Inc.

Date Written: July 1, 2020


This paper demonstrates how deep learning can be used to price and calibrate models of credit risk. Deep neural networks can learn structural and reduced-form models with high degrees of accuracy. For complex credit risk models, whose closed-form solutions are not available, deep learning offers a conceptually simple and more efficient alternative solution. We propose an approach that combines deep learning with the unscented Kalman filter to calibrate credit risk models on historical data, which attains an in-sample R-squared of 98.5 percent for the reduced-form model and 95 percent for the structural model.

Keywords: Deep Learning, Machine Learning, Credit Risk Modeling, Default Risk, Sovereign Risk, Neural Networks

JEL Classification: G10, G12, G17

Suggested Citation

Manzo, Gerardo and Qiao, Xiao, Deep Learning Credit Risk Modeling (July 1, 2020). Forthcoming Journal of Fixed Income, Available at SSRN: or

Gerardo Manzo (Contact Author)

Independent ( email )

New York, NY 10018
United States

Xiao Qiao

School of Data Science, City University of Hong Kong ( email )

Hong Kong

Paraconic Technologies US Inc. ( email )

New York, NY
United States


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

Paper statistics

Abstract Views
PlumX Metrics