Zhongyun Li
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Zhongyun Li has a total of 8 co-authors. Here are the top co-authors by number of shared publications:
- Jiaxing Wang
College of Economics and Management, Northwest A&F University, Xianyang, China
4 shared publications - Guoquan Liu
Institute of FinTech, Fudan University, Shanghai, China
2 shared publications Xiaobo Xu
International Business School Suzhou, Xian Jiaotong Liverpool University, Suzhou, China
2 shared publicationsChenrui Li
1 shared publication- Haoming Yan
College of Economics & Management, Northwest A&F University, Yangling, Shaanxi, China
1 shared publication Kaiyue Liu
1 shared publication
Latest Publications
- Open Access
- 15 April 2026
- Published by Elsevier in SSRN Electronic Journal
Pre-announcement trading pressure, firm size, and late-stage price adjustment: Evidence from FTSE 100 financial report announcements
This paper examines whether a pre-announcement trading-pressure proxy, motivated in part by the overconfidence literature, predicts announcement-period price adjustment in the FTSE 100. Using 2,080 firm-event observations from 2014 to 2023, I construct a reduced-form composite from standardized abnormal volume, a return-sign-based directional trading indicator, and a turnover-volatility measure over the 30 trading days preceding each scheduled financial-report-related disclosure and estimate firm- and year-fixed-effects models for cumulative abnormal returns (CAR), announcement-period volatility (VOL), and a convergence-based delay measure (Delay). The full-sample average association is weak. The central result is conditional: the interaction between the proxy and firm size is negative and significant in the Delay equation. The size-conditioned effect emerges at 80% and 90% convergence thresholds but not at 50% or 70%, indicating a late-stage rather than general speed effect. Decomposition tests show that turnover volatility explains most of the composite's return and volatility content but does not account for the Delay interaction. Overall, the evidence suggests a size-conditioned pattern in final-stage price convergence, with information intermediation, liquidity provision, and arbitrage capacity plausibly more concentrated among the very largest constituents.
- 15 April 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessPre-announcement trading pressure, firm size, and late-stage price adjustment: Evidence from FTSE 100 financial report announcements
This paper examines whether a pre-announcement trading-pressure proxy, motivated in part by the overconfidence literature, predicts announcement-period price adjustment in the FTSE 100. Using 2,080 firm-event observations from 2014 to 2023, I construct a reduced-form composite from standardized abnormal volume, a return-sign-based directional trading indicator, and a turnover-volatility measure over the 30 trading days preceding each scheduled financial-report-related disclosure and estimate firm- and year-fixed-effects models for cumulative abnormal returns (CAR), announcement-period volatility (VOL), and a convergence-based delay measure (Delay). The full-sample average association is weak. The central result is conditional: the interaction between the proxy and firm size is negative and significant in the Delay equation. The size-conditioned effect emerges at 80% and 90% convergence thresholds but not at 50% or 70%, indicating a late-stage rather than general speed effect. Decomposition tests show that turnover volatility explains most of the composite's return and volatility content but does not account for the Delay interaction. Overall, the evidence suggests a size-conditioned pattern in final-stage price convergence, with information intermediation, liquidity provision, and arbitrage capacity plausibly more concentrated among the very largest constituents.
- 22 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Engineering Management
- Vol. 72 (00189391) , 3262-3278
A Multitask Learning Aspect-Based Model for Product Defect Detection
Product defect detection (PDD) is a critical industrial big data application that leverages social media text mining to identify product defects. However, existing PDD approaches face significant challenges: heavy reliance on expert-driven aspect term extraction (ATE) limits scalability and increases costs, and they fail to capture detailed defect information like affected aspects, defect causes, and severity. To address these limitations, this study proposes a novel multi-task learning framework that integrates ATE and aspect-level defect detection. By enhancing ATE with semantic augmentation, this model reduces expert dependence and scales effectively. Experiments on automotive datasets show the proposed method improves the F1 score by 5% (to 0.88) over benchmark models. Furthermore, ablation studies reveal a 36.5% F1 score improvement for the ATE component. The proposed method outperforms benchmark PDD models and provides more fine-grained defect information, enabling a more comprehensive understanding and supporting precise decision-making in quality management.
- 22 July 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Engineering Management
- Vol. 72 (00189391) , 3262-3278
A Multitask Learning Aspect-Based Model for Product Defect Detection
Product defect detection (PDD) is a critical industrial big data application that leverages social media text mining to identify product defects. However, existing PDD approaches face significant challenges: heavy reliance on expert-driven aspect term extraction (ATE) limits scalability and increases costs, and they fail to capture detailed defect information like affected aspects, defect causes, and severity. To address these limitations, this study proposes a novel multi-task learning framework that integrates ATE and aspect-level defect detection. By enhancing ATE with semantic augmentation, this model reduces expert dependence and scales effectively. Experiments on automotive datasets show the proposed method improves the F1 score by 5% (to 0.88) over benchmark models. Furthermore, ablation studies reveal a 36.5% F1 score improvement for the ATE component. The proposed method outperforms benchmark PDD models and provides more fine-grained defect information, enabling a more comprehensive understanding and supporting precise decision-making in quality management.
- Open Access
- 13 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 150-156
From Innovation to Implementation: Factors Driving Chinese Farmers' Adoption Intentions for Laser Weeding Robots
- 13 June 2025
- Published by Association for Computing Machinery (ACM)
- p. 150-156
Open AccessFrom Innovation to Implementation: Factors Driving Chinese Farmers' Adoption Intentions for Laser Weeding Robots
- 1 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Engineering Management
- Vol. 72 (00189391) , 1924-1938
Leveraging Internet-Sourced Text Data for Financial Analytics in Supply Chain Finance: A Large Language Model-Enhanced Text Mining Workflow
In the era of artificial intelligence and fintech, improving the efficiency of financial analysis is essential for financial service providers. This article proposes a novel large language model-enhanced text mining workflow that leverages Internet-sourced text information to efficiently analyze supply chain finance business without requiring programming skills. We conduct a case study on the Chinese market for new energy buses—a rapidly growing sector due to government incentives and the push for sustainable urban transportation—using data from bidding websites and financial statements. The experimental results demonstrate that our LLM-enhanced workflow outperforms traditional methods, showcasing increased efficiency and practicality, especially for non-programming employees in supply chain financial services.
- 1 January 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Engineering Management
- Vol. 72 (00189391) , 1924-1938
Leveraging Internet-Sourced Text Data for Financial Analytics in Supply Chain Finance: A Large Language Model-Enhanced Text Mining Workflow
In the era of artificial intelligence and fintech, improving the efficiency of financial analysis is essential for financial service providers. This article proposes a novel large language model-enhanced text mining workflow that leverages Internet-sourced text information to efficiently analyze supply chain finance business without requiring programming skills. We conduct a case study on the Chinese market for new energy buses—a rapidly growing sector due to government incentives and the push for sustainable urban transportation—using data from bidding websites and financial statements. The experimental results demonstrate that our LLM-enhanced workflow outperforms traditional methods, showcasing increased efficiency and practicality, especially for non-programming employees in supply chain financial services.