RT Journal Article SR Electronic T1 Deep Learning Credit Risk Modeling JF The Journal of Fixed Income FD Institutional Investor Journals SP jfi.2021.1.121 DO 10.3905/jfi.2021.1.121 A1 Gerardo Manzo A1 Xiao Qiao YR 2021 UL https://pm-research.com/content/early/2021/08/02/jfi.2021.1.121.abstract AB This article 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 with no closed-form solutions available, deep learning offers a conceptually simple and more efficient alternative solution. This article proposes an approach that combines deep learning with the unscented Kalman filter to calibrate credit risk models based on historical data; this strategy attains an in-sample R-squared of 98.5% for the reduced-form model and 95% for the structural model.TOPICS: Credit risk management, big data/machine learning, quantitative methods, statistical methodsKey Findings▪ Neural networks can approximate solutions to credit risk models, precisely capturing the relationship between model inputs and credit spreads.▪ Compared to standard techniques, the approximate solutions are more computationally efficient.▪ Neural networks can be used to accurately calibrate structural and reduced-form models of credit risk.