Author(s)

HaiLong Liao

Affiliation(s)

School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing 100049, China.

Corresponding Author

HaiLong Liao

ABSTRACT

This research focuses on the application of the DeepSeek large - scale model in intelligent stock selection in the A - share market. It constructs a multi - dimensional factor analysis framework that integrates reinforcement learning and a mixture - of - experts architecture. Through empirical research, the advantages of this model in terms of return acquisition and risk control are verified, providing new intelligent strategies for A - share investment and promoting technological innovation and development in the capital market.

KEYWORDS

DeepSeek large - scale model; Intelligent stock selection in the A - share market; Factor system; Empirical research

CITE THIS PAPER

HaiLong Liao. A-share intelligent stock selection strategy based on the DeepSeek large model: Technical routes, factor systems, and empirical research. Eurasia Journal of Science and Technology. 2025, 7(2): 7-13. DOI: https://doi.org/10.61784/ejst3070.

REFERENCES

[1] Eugene F, Fama, Kenneth R, French. Fama - French Three - Factor Model and Its Application. 1993. Chinese Software Developer Network. Retrieved from https://blog.csdn.net/hzk427/article/details/104187847 

[2] Liu Yaning. ChatGPT Consumes Over 500,000 kWh of Electricity per Day. How Energy - Intensive Are AI Large - Scale Models? Titanium Media. 2024. Retrieved from https://baijiahao.baidu.com/s?id = 1793597173423030217&wfr = spider&for = pc

[3] Qiao Nan. Decoding the Innovation Path of DeepSeek: The Evolution Roadmap of Three Generations of Models. 2025. Retrieved from https://news.sina.cn/ai/2025 - 02 - 09/detail - ineiwqye4694847.d.html

[4] Liao Hailong. DeepSeek Large - Scale Model: Technical Analysis and Development Prospect. Journal of Computer Science and Electrical Engineering, 2025, 7(1): 33-37. DOI: https://doi.org/10.61784/jcsee3035

[5] Lieyan Benniu. DeepSeek: Algorithm Innovation Leads to an AI Computing Power Boom and Usher in a New Era of Intelligent Technology. 2025. Retrieved from http://mp.weixin.qq.com/s?__biz = Mzg2MTE3NDA2Mg==&mid = 2247494898&idx = 1&sn = b99504080dcd03e5a6809136486d5bde&scene = 0

[6] Wang Houdong. The Important Impact of DeepSeek on the Global Development of Artificial Intelligence. 2025. Retrieved from http://mp.weixin.qq.com/s?__biz = MjM5MjgzMzg1Nw==&mid = 2453991839&idx = 1&sn = 74a9185fb47c86c8deb4939df9b81cd5&scene = 0

[7] Demon Akana. Introduction to the Technical Principles of Deepseek's MLA Technology. Chinese Software Developer Network. 2025. Retrieved from https://blog.csdn.net/bestpasu/article/details/145539423 

[8] Baitai Laoren. DeepSeekMoE Architecture. Chinese Software Developer Network. 2024. Retrieved from https://blog.csdn.net/weixin_41429382/article/details/144701990 

[9] Zhihong Shao, Peiyi Wang, Qihao Zhu, et al. DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models. 2024.  DOI: https://doi.org/10.48550/arXiv.2402.03300. Retrieved from https://arxiv.org/pdf/2402.03300 

[10] Le Tutu. In - Depth Analysis of Other Potential Defects of DeepSeek. Eastmoney Wealth Channel. 2025. Retrieved from https://caifuhao.eastmoney.com/news/20250202221538225857610