Construction of High-frequency Large Capital Flow Factor and Its Predictive Effect Comparison on Excess Returns of A-share ESG Portfolios

Xinyi He

doi:10.54691/2pkzhy10

Authors

Xinyi He Hong Kong University of Science and Technology, Hong Kong, China

DOI:

https://doi.org/10.54691/2pkzhy10

Keywords:

High-frequency factor, large capital flow, ESG portfolio, excess return, return prediction, A-share market.

Abstract

Against the background of the rapid development of ESG investment and quantitative factor trading in China’s A-share market, institutional large capital flow is a core proxy variable of smart money trading behavior, which contains effective predictive information of stock price changes. This paper constructs a high-frequency large capital flow (LCF) factor based on intraday trading data of A-shares, and systematically explores its differential predictive effects on the excess returns of ESG portfolios with different rating levels. Based on the monthly panel data of A-share listed companies from 2020 to 2024, this study divides samples into high, medium and low ESG portfolios, and adopts Fama-Macbeth regression and portfolio sorting methods to test the predictive ability of the LCF factor on portfolio excess returns. The empirical results show that the constructed high-frequency LCF factor has significant positive predictive power on the excess returns of A-share ESG portfolios; specifically, the predictive effect is the most prominent for high ESG-rated portfolios, with a monthly excess return increment of 0.87% and a significant t-statistic at the 1% level. Further mechanism tests indicate that high ESG enterprises have lower information asymmetry and stronger institutional investor preference, which amplifies the price discovery efficiency of large capital flow signals. This paper innovatively combines high-frequency capital flow characteristics with ESG portfolio pricing, and provides empirical evidence and quantitative strategy reference for ESG factor investment and smart money trading decision-making in China’s capital market.

Downloads

Download data is not yet available.

References

[1] Yu, H., Liang, C., Liu, Z., & Cui, X. (2025). The effectiveness of news‐based ESG sentiment for predicting stock returns: Evidence from China. Accounting & Finance, 65(3), 2724–2732.

[2] Octaviano, M. K., & Seong, J. T. (2026). From risk to reward: Learning ESG return dynamics in Chinese equities. AIMS Mathematics, 11(6), 16448–16478.

[3] Chen, B. (2025). Can machine learning uncover ESG alpha in the Chinese A-share market? An “alpha illusion” case study. Finance Research Letters, 109462.

[4] Liu, X., Xiao, X., Shen, S., & Xu, Z. (2026). War amongst ESG ratings in China: A battle of stock return predictability. Green and Low-Carbon Economy, 4(2), 208–221.

[5] Sun, E. W., Kruse, T., & Yu, M. T. (2014). High frequency trading, liquidity, and execution cost. Annals of Operations Research, 223(1), 403–432.

[6] Chen, S., Zhang, J., & Ge, L. (2026). Predicting ESG ratings with recurrent neural network models: Evidence from China’s A-share listed companies. Computational Economics, 1–23.

[7] Chen, M., & Garriott, C. (2020). High-frequency trading and institutional trading costs. Journal of Empirical Finance, 56, 74–93.

[8] Van Heijningen, K. (2019). The impact of ESG factor materiality on stock performance of firms. Erasmus Platform for Sustainable Value Creation working paper.

[9] Sun, Y., Jiao, D., Yang, Y., Peng, Y., & Hu, S. (2025). Impact of ESG preferences on investors in China’s A-share market. International Journal of Financial Studies, 13(4), 191.

[10] Wang, Z., Liao, K., & Zhang, Y. (2022). Does ESG screening enhance or destroy stock portfolio value? Evidence from China. Emerging Markets Finance and Trade, 58(10), 2927–2941.

[11] Zhan, M., Zheng, L., & Da, J. (2025, March). Research on ESG excess returns based on machine learning and cloud computing. In 2025 International Conference on Electrical Drives, Power Electronics & Engineering (EDPEE) (pp. 555–559). IEEE.