Research

Abstract: We conduct a comprehensive study on momentum spillovers in the Chinese stock market using various types of economic linkages. We find that the news co-mention momentum spillover is significantly stronger compared to other forms of momentum spillovers. Using spanning tests and Fama-MacBeth regressions, we further show that the news co-mention momentum spillover unifies all different forms of momentum spillover effects in the Chinese stock market. Notably, the analyst co-coverage momentum spillover effect, which is the dominant species in the US stock market, is subsumed by the news co-mention momentum spillover effect in the Chinese stock market. We further explore the differences in the information content of links implied by news co-mentioning and other proxies. We suggest that the dominance of news co-mention momentum spillover over others can be attributed to two primary factors: comprehensive information and prompt updates.

Abstract:  We propose a similarity criterion for stocks based on their Euclidean distance over a set of attention-related variables to capture investors' memory of stocks. We discover that the average past return of similar-attention stocks exhibits a significantly strong, negative predictive power for the future return of a focal stock, suggesting a cross-stock reversal effect. A long-short strategy based on cross-stock reversal generates a monthly return of 2.23% and an annualized Sharpe ratio of 1.63. This effect cannot be explained by common factors including the short-term reversal. We further show that the cross-stock reversal effect comes from investors’ memory-based belief and irrational trading from cross-extrapolating past performances of similar-attention stocks. The predictability is related to the investor clientele since it performs poorly among shares with more professional investors. 

Presentations: The 5th Greater China Area Finance Conference;  2025 Dishui Lake International Conference in Finance; 2025 China FinTech Research Conference; 2025 Asian FA (Scheduled); 2025 CICF (Scheduled); 

Abstract: This paper investigates how different structures of information presentation in financial news affect investors' attention efficiency via processing costs and subsequently influence market dynamics. We decompose news-implied links to two types with distinct co-mentioning structures: leader-follower (LF) and peer (PE) links, and empirically find that LF links are significantly different from PE links regarding information diffusion, resulting in faster attention spillovers (183%), as measured by abnormal search volume indices (ASVI), weaker cross-stock momentum (52% less per month), as shown by portfolio analysis, and quicker cross-firm price discovery (12%), compared to PE links. By excluding potential attention level effects and controlling for other factors that could affect information processing costs, we conclude that co-mentioning structures influence attention efficiency independently through LF links' directed networks, and journalists have the discretion to improve investors' attention efficiency.