Topic: Predicting Corporate Bond Returns:Merton Meets Machine Learning

Lecturer: Prof. Fuwei JIANG, Central University of Finance and Economics

Time: 14:30-16:00 p.m. May 5th,2023 (Fri.)

Venue: B423, Zhixin Building, Central Campus

Abstract:We investigate the return predictability of corporatebonds using big data and machine learning. We find thatmachine learning models substantially improve the out-ofsample performance of stock and bond characteristics inpredicting future bond returns. We also find a significantimprovement in the performance of machine learningmodels when imposing a theoretically motivated eco .nomic structure from the Merton model, compared to thereduced-form approach without restrictions. Overall, ourwork highlights the importance of explicitly imposing thedependence between expected bond and stock returns viamachine learning and Merton model when investigatingexpected bond returns