arXiv q-fin· Derun Zhong, Guolian Wang, Shiqing Wang, Hongjie Feng·· 5 hr agoAI Score61
Empirical Test of Cross-Bond Panel Learning for Next-Day Price Prediction of Chinese Convertible Bonds: Deep Models Fail to Outperform the Random Walk
Cross-Bond Panel Learning for Next-Day Price Forecasting of Chinese Convertible Bonds: A Reality Check
AI summary
This arXiv paper conducts a rigorous empirical test of the next-day price prediction capability of cross-bond panel learning for Chinese convertible bonds. Based on 73,806 daily observations across 100 A-share convertible bonds from 2022 to 2025, it compares a random walk benchmark against cross-bond neural network models on a held-out set of 29,514 observations.
Source:arXiv q-fin · arxiv.org