Interpretable and Arbitrage-free Deep Learning for Corporate Bond Pricing
Speaker：FENG Guanhao（Assistant Professor, City University of Hong Kong）
Host：LIU Yanchu, Associate Professor, Lingnan College
Time and Date：16:00-18:30, Dec. 17, 2021
Language: English + Chinese
Abstract：
This paper combines asset pricing theory with deep learning for pricing the cross section of corporate bonds. The proposed deep learning model can flexibly introduce the well-established factors and provide us with latent factors that are not subsumed in those existing factors. The generated deep factors are tradable long-short portfolios based on a large number of bond and equity characteristics and hence are economically more interpretable. We show empirically that our deep learning factor model improves the asset pricing performance on various corporate bond portfolios over standard factor models and recommends bond investment portfolios that outperform the leading corporate bond strategies.
Profile of the speaker:
Guanhao (Gavin) Feng is an assistant professor of business statistics at the City University of Hong Kong. He is also the program leader of MSc. in Business Data Analytics (Quantitative Analysis for Business stream), a faculty affiliate at the School of Data Science, and a Co-PI in the Lab for AI-Powered FinTech. Gavin
