This is the authors' abstract. We don't add key points for this paper.
• Among the first to apply machine learning models to Chinese convertible bond market. • We design 9 factor types with over 27 factors for Chinese convertible bonds. • Portfolio results show XGB model outperforms others in Chinese convertible bond market. Using data from 2018 to 2023 on Chinese convertible bonds and their underlying stocks, we examine the cross-sectional returns by constructing factor-based portfolios across nine categories and employing various machine learning models. Our results indicate that extreme gradient boosting (XGB) consistently outperforms alternative methods, demonstrating its superior ability to capture nonlinearities and factor interactions. Based on machine learning predictions, we further design long–short portfolios that generate economically significant gains, with the XGB-based strategy achieving an average monthly return of 1.52%, statistically significant at the 1% level. This study contributes to the growing literature applying machine learning to convertible bonds in the Chinese market and highlights the practical value of these methods for designing profitable investment strategies, with potential implications for derivative markets more broadly.
Answers are grounded in this paper and cite their sources.
No takes yet. Share an insight, caveat, or question.
Luo et al. (2025) studied this question.
Synapse has enriched 8 closely related papers on similar clinical questions. Consider them for comparative context: