Similar articles

Abstract: Traditional quantitative investment research is encountering diminishing returns alongside rising labor and time costs. To overcome these challenges, we introduce the large investment model (LIM), a novel research paradigm designed to enhance both performance and efficiency at scale. LIM employs end-to-end learning and universal modeling to create an upstream foundation model, which is capable of autonomously learning comprehensive signal patterns from diverse financial data spanning multiple exchanges, instruments, and frequencies. These "global patterns" are subsequently transferred to downstream strategy modeling, optimizing performance for specific tasks. We detail the system architecture design of LIM, address the technical challenges inherent in this approach, and outline potential directions for future research.

投资大模型

[1]Ahlberg JH, Nilson EN, Walsh JL, 1967. The Theory of Splines and Their Applications. Academic Press, New York, NY, USA.

[2]Ait-Sahalia Y, Saglam M, 2024. High frequency market making: the role of speed. J Econometr, 239(2):105421.

[3]Aldridge I, 2013. High-Frequency Trading: a Practical Guide to Algorithmic Strategies and Trading Systems (2^nd Ed.). Wiley, Hoboken, USA.

[4]Angles R, Gutierrez C, 2008. Survey of graph database models. ACM Comput Surv, 40(1):1-39.

[5]Bengio Y, Courville A, Vincent P, 2013. Representation learning: a review and new perspectives. IEEE Trans Patt Anal Mach Intell, 35(8):1798-1828.

[6]Black F, Scholes M, 1973. The pricing of options and corporate liabilities. J Pol Econ, 81(3):637-654.

[7]Bodie Z, Kane A, Marcus AJ, 2014. Investments (10^th Ed.). McGraw-Hill Education, New York, USA.

[8]Brown P, Walsh D, Yuen A, 1997. The interaction between order imbalance and stock price. Pacif-Bas Fin J, 5(5):539-557.

[9]Brown T, Mann B, Ryder N, et al., 2020. Language models are few-shot learners. Advances in Neural Information Processing Systems, 33:1877-1901.

[10]Cao SS, Jiang W, Lei LG, et al., 2024. Applied AI for finance and accounting: alternative data and opportunities. Pacif-Bas Fin J, 84:102307.

[11]Carbone P, Katsifodimos A, Ewen S, et al., 2015. Apache Flink: stream and batch processing in a single engine. IEEE Datab Eng Bull, 36:28-33.

[12]Chan KC, 1988. On the contrarian investment strategy. J Bus, 61(2):147-163.

[13]Chen Q, Francis J, Jiang W, 2005. Investor learning about analyst predictive ability. J Account Econ, 39(1):3-24.

[14]Chen T, Chen W, Du L, 2021. An empirical study of financial factor mining based on gene expression programming. 4^th Int Conf on Advanced Electronic Materials, Computers and Software Engineering, p.1113-1117.

[15]Chincarini L, Kim D, 2022. Quantitative Equity Portfolio Management: an Active Approach to Portfolio Construction and Management (2^nd Ed.). McGraw-Hill Education, New York, USA.

[16]Cho P, Park JH, Song JW, 2021. Equity research report-driven investment strategy in Korea using binary classification on stock price direction. IEEE Access, 9:46364-46373.

[17]Coletta A, Jerome J, Savani R, et al., 2023. Conditional generators for limit order book environments: explainability, challenges, and robustness. Proc 4^th ACM Int Conf on AI in Finance, p.27-35.

[18]Das A, Kong W, Sen R, et al., 2024. A decoder-only foundation model for time-series forecasting. Proc 41^st Int Conf on Machine Learning, p.10148-10167.

[19]Das V, 2015. Learning Redis. Packt Publishing Ltd., Birmingham, UK.

[20]Daumé HIII, 2007. Frustratingly easy domain adaptation. Proc 45^th Annual Meeting of the Association of Computational Linguistics, p.256-263.

[21]Dempster M, Kanniainen J, Keane J, et al., 2018. High-Performance Computing in Finance: Problems, Methods, and Solutions (1^st Ed.). Chapman and Hall/CRC, New York, NY, USA.

[22]Dessaint O, Foucault T, Fresard L, 2024. Does alternative data improve financial forecasting? The horizon effect. J Fin, 79(3):2237-2287.

[23]Dixon M, 2022. The book of alternative data: a guide for investors, traders and risk managers. Quant Fin, 22(8):1427-1428.

[24]Dixon W, Yuen K, 1974. Trimming and winsorization: a review. Stat Hefte, 15:157-170.

[25]Dunning T, Friedman B, Loukides M, et al., 2014. Time Series Databases: New Ways to Store and Access Data. O'Reilly Media, Incorporated.

[26]Engelberg J, McLean RD, Pontiff J, et al., 2023. Do cross-sectional predictors contain systematic information? J Fin Quant Anal, 58(3):1172-1201.

[27]Fabozzi F, Mann S, Fabozzi F, 2021. The Handbook of Fixed Income Securities (9^th Ed.). McGraw-Hill Education, New York, USA.

[28]Fama EF, French KR, 1992. The cross-section of expected stock returns. J Fin, 47(2):427-465.

[29]Fang Y, Ren K, Liu W, et al., 2021. Universal trading for order execution with Oracle policy distillation. Proc AAAI Conf on Artificial Intelligence, 35:107-115.

[30]Fang Y, Tang Z, Ren K, et al., 2023. Learning multi-agent intention-aware communication for optimal multi-order execution in finance. Proc 29^th ACM SIGKDD Conf on Knowledge Discovery and Data Mining, p.4003-4012.

[31]Feng F, He X, Wang X, et al., 2019. Temporal relational ranking for stock prediction. ACM Trans Inform Syst, 37(2):1-30.

[32]Feng F, Wang X, He X, et al., 2021. Time horizon-aware modeling of financial texts for stock price prediction. Proc 2^nd ACM Int Conf on AI in Finance, p.1-8.

[33]Gastineau GL, 2001. Exchange-traded funds: an introduction. J Portf Manag, 27(3):88-96.

[34]Goodfellow I, Bengio Y, Courville A, 2016. Deep Learning. MIT Press, Cambridge, USA.

[35]Gupta T, Leung E, Roscovan V, 2022. Consumer spending and the cross-section of stock returns. J Portf Manag, 48(7):117-137.

[36]Han J, E H, Le G, et al., 2011. Survey on NoSQL database. 6^th Int Conf on Pervasive Computing and Applications, p.363-366.

[37]Han S, Pool J, Tran J, et al., 2015. Learning both weights and connections for efficient neural networks. Proc 29^th Int Conf on Neural Information Processing Systems, p.1135-1143.

[38]Hastie T, Tibshirani R, Friedman J, 2009. The Elements of Statistical Learning: Data Mining, Inference, and Prediction (2^nd Ed.). Springer, New York, USA.

[39]Ho T, Stoll HR, 1981. Optimal dealer pricing under transactions and return uncertainty. J Fin Econ, 9(1):47-73.

[40]Hochreiter S, Schmidhuber J, 1997. Long short-term memory. Neur Comput, 9(8):1735-1780.

[41]Howard J, Ruder S, 2018. Universal language model fine-tuning for text classification. Proc 56^th Annual Meeting of the Association for Computational Linguistics, p.328-339.

[42]Hu EJ, Shen Y, Wallis P, et al., 2021. LoRA: low-rank adaptation of large language models. https://arxiv.org/abs/2106.09685

[43]Hu Z, Liu W, Bian J, et al., 2018. Listening to chaotic whispers: a deep learning framework for news-oriented stock trend prediction. Proc 11^th ACM Int Conf on Web Search and Data Mining, p.261-269.

[44]Hull JC, 2021. Options, Futures and Other Derivatives (11^th Ed.). Pearson, Hoboken, NJ, USA.

[45]Jacob B, Kligys S, Chen B, et al., 2018. Quantization and training of neural networks for efficient integer-arithmetic-only inference. IEEE/CVF Conf on Computer Vision and Pattern Recognition, p.2704-2713.

[46]Jiao X, Li Z, Xu C, et al., 2023. Microstructure-empowered stock factor extraction and utilization.

[47]Kakushadze Z, 2016. 101 formulaic alphas. Wilmott, 2016(84):72-81.

[48]Kim AY, Tse Y, Wald JK, 2016. Time series momentum and volatility scaling. J Fin Mark, 30:103-124.

[49]Kou Y, Wang CY, Ye Q, 2013. An empirical study of calendar spread arbitrage based on high-frequency data: the case of CSI 300 index futures. Int Conf on Management Science and Engineering, p.1604-1609.

[50]Krauss C, 2017. Statistical arbitrage pairs trading strategies: review and outlook. J Econ Surv, 31(2):513-545.

[51]Kwan C, 1999. A note on market-neutral portfolio selection. J Bank Fin, 23(5):773-800.

[52]Lewis P, Perez E, Piktus A, et al., 2020. Retrieval-augmented generation for knowledge-intensive NLP tasks. Advances in Neural Information Processing Systems, 33:9459-9474.

[53]Li Y, Wang T, Sun B, et al., 2022. Detecting the lead–lag effect in stock markets: definition, patterns, and investment strategies. Fin Innov, 8(1):51.

[54]Li Y, Huang Y, Ildiz ME, et al., 2024. Mechanics of next token prediction with self-attention. Int Conf on Artificial Intelligence and Statistics, p.685-693.

[55]Liew JKS, Budavari T, Kang Z, et al., 2020. Pairs trading strategy with geolocation data: the battle between Under Armour and Nike. J Fin Data Sci, 2(1):126-143.

[56]Lin S, Beling PA, 2021. An end-to-end optimal trade execution framework based on proximal policy optimization. Proc 29^th Int Joint Conf on Artificial Intelligence (Special Track on AI in FinTech), p.4548-4554.

[57]Liu T, Roberts S, Zohren S, 2023. Deep inception networks: a general end-to-end framework for multi-asset quantitative strategies.

[58]Lo AW, 2010. Hedge Funds: an Analytic Perspective (Revised and Expanded Edition). Princeton University Press, Princeton, NJ, USA. http://www.jstor.org/stable/j.ctt7rq28

[59]Lo AW, Mamaysky H, Wang J, 2000. Foundations of technical analysis: computational algorithms, statistical inference, and empirical implementation. J Fin, 55(4):1705-1765.

[60]MacKinlay AC, 1997. Event studies in economics and finance. J Econ Liter, 35(1):13-39.

[61]Madhavan A, 2000. Market microstructure: a survey. J Fin Mark, 3(3):205-258.

[62]Mao A, Mohri M, Zhong Y, 2023. Cross-entropy loss functions: theoretical analysis and applications. Proc 40^th Int Conf on Machine Learning, p.23803-23828.

[63]Nagy P, Frey S, Sapora S, et al., 2023. Generative AI for end-to-end limit order book modelling: a token-level autoregressive generative model of message flow using a deep state space network. 4^th ACM Int Conf on AI in Finance, p.91-99.

[64]Nie Y, Nguyen NH, Sinthong P, et al., 2023. A time series is worth 64 words: long-term forecasting with Transformers. https://arxiv.org/abs/2211.14730

[65]OpenAI, Achiam J, Adler S, et al., 2024. GPT-4 technical report. https://arxiv.org/abs/2303.08774

[66]Paatela A, Noschis E, Hameri AP, 2017. Abnormal stock returns using supply chain momentum and operational financials. J Portf Manag, 43:50-60.

[67]Pan SJ, Yang Q, 2010. A survey on transfer learning. IEEE Trans Knowl Data Eng, 22(10):1345-1359.

[68]Păuna C, 2019. Additional limit conditions for breakout trading strategies. Inform Econ, 23(2):25-33.

[69]Penman SH, 2013. Financial Statement Analysis and Security Valuation (5^th Ed.). McGraw-Hill Education, New York, USA.

[70]Pfeiffer J, Ruckle A, Gurevych I, 2021. AdapterFusion: non-destructive task composition for transfer learning. Proc 16^th Conf of the European Chapter of the Association for Computational Linguistics, p.487-503.

[71]Qin M, Sun S, Zhang W, et al., 2024. EarnHFT: efficient hierarchical reinforcement learning for high frequency trading. Proc 38^th AAAI Conf on Artificial Intelligence and 36^th Conf on Innovative Applications of Artificial Intelligence and 14^th Symp on Educational Advances in Artificial Intelligence, p.14669-14676.

[72]Salloum S, Dautov R, Chen X, et al., 2016. Big data analytics on Apache Spark. Int J Data Sci Anal, 1(3):145-164.

[73]Saputra MYE, Arief SN, Wijayaningrum VN, et al., 2024. Real-time server monitoring and notification system with Prometheus, Grafana, and Telegram integration. ASU Int Conf in Emerging Technologies for Sustainability and Intelligent Systems, p.1808-1813.

[74]Sawhney R, Agarwal S, Wadhwa A, et al., 2020. Spatiotemporal hypergraph convolution network for stock movement forecasting. IEEE Int Conf on Data Mining, p.482-491.

[75]Sawhney R, Agarwal S, Wadhwa A, et al., 2021. Stock selection via spatiotemporal hypergraph attention network: a learning to rank approach. Proc AAAI Conf on Artificial Intelligence, 35(1):497-504.

[76]Sharpe WF, 1966. Mutual fund performance. J Bus, 39(1):119-138.

[77]Sheikh A, 1996. Barra's risk models. Barra Res Insights, 1:1-24.

[78]Shleifer A, Vishny RW, 1997. The limits of arbitrage. J Fin, 52(1):35-55.

[79]Shvachko K, Kuang H, Radia S, et al., 2010. The Hadoop distributed file system. IEEE 26^th Symp on Mass Storage Systems and Technologies, p.1-10.

[80]Sorge M, 2004. Stress-testing financial systems: an overview of current methodologies. BIS Working Paper, No. 165.

[81]Stan CS, Pandelica AE, Zamfir VA, et al., 2019. Apache Spark and Apache Ignite performance analysis. 22^nd Int Conf on Control Systems and Computer Science, p.726-733.

[82]Stonebraker M, 2010. SQL databases v. NoSQL databases. Commun ACM, 53(4):10-11.

[83]Sun S, Xue W, Wang R, et al., 2022. DeepScalper: a risk-aware reinforcement learning framework to capture fleeting intraday trading opportunities. Proc 31^st ACM Int Conf on Information & Knowledge Management, p.1858-1867.

[84]Sun Y, Liu L, Xu Y, et al., 2024. Alternative data in finance and business: emerging applications and theory analysis. Fin Innov, 10(1):127.

[85]Tetlock PC, Saar-Tsechansky M, Macskassy S, 2008. More than words: quantifying language to measure firms' fundamentals. J Fin, 63(3):1437-1467.

[86]Toth B, Eisler Z, Lillo F, et al., 2012. How does the market react to your order flow? Quant Fin, 12(7):1015-1024.

[87]Vaswani A, Shazeer N, Parmar N, et al., 2017. Attention is all you need. Advances in Neural Information Processing Systems, 30:5998-6008.

[88]Wafi AS, Hassan H, Mabrouk A, 2015. Fundamental analysis models in financial markets—review study. Proc Econ Fin, 30:939-947.

[89]Wang J, Zhang Y, Tang K, et al., 2019. AlphaStock: a buying-winners-and-selling-losers investment strategy using interpretable deep reinforcement attention networks. Proc 25^th ACM SIGKDD Int Conf on Knowledge Discovery & Data Mining, p.1900-1908.

[90]Wang L, Zhang X, Su H, et al., 2024. A comprehensive survey of continual learning: theory, method and application. IEEE Trans Patt Anal Mach Intell, 46(8):5362-5383.

[91]Wang L, Chen S, Jiang L, et al., 2025. Parameter-efficient fine-tuning in large language models: a survey of methodologies. Artif Intell Rev, 58:227.

[92]Wang S, Yuan H, Zhou L, et al., 2025. Alpha-GPT: human–AI interactive alpha mining for quantitative investment.

[93]Wang Z, Huang B, Tu S, et al., 2021. DeepTrader: a deep reinforcement learning approach for risk-return balanced portfolio management with market conditions embedding. Proc AAAI Conf on Artificial Intelligence, 35(1):643-650.

[94]Wei Z, Dai B, Lin D, 2023. E2EAI: end-to-end deep learning framework for active investing. Proc 4^th ACM Int Conf on AI in Finance, p.55-63.

[95]Xia SQ, Simonian J, 2021. Measuring investment skill in multi-asset strategies: an empirical study of the information coefficient as weighted rank correlation. J Portf Manag, 47(4):135-144.

[96]Xie X, Liu H, Hou W, et al., 2023. A brief survey of vector databases. 9^th Int Conf on Big Data and Information Analytics, p.364-371.

[97]Xu Y, Cohen SB, 2018. Stock movement prediction from Tweets and historical prices. Proc 56^th Annual Meeting of the Association for Computational Linguistics, p.1970-1979.

[98]Yang W, Wei Y, Wei H, et al., 2023. Survey on explainable AI: from approaches, limitations and applications aspects. Human-Centr Intell Syst, 3:161-188.

[99]Yin S, Fu C, Zhao S, et al., 2024. A survey on multimodal large language models. Nat Sci Rev, 11(12).

[100]Yosinski J, Clune J, Bengio Y, et al., 2014. How transferable are features in deep neural networks? Proc 28^th Int Conf on Neural Information Processing Systems, p.3320-3328.

[101]Yu H, Hao X, Wu L, et al., 2023. Eye in outer space: satellite imageries of container ports can predict world stock returns. Human Soc Sci Commun, 10(1):1-16.

[102]Zhang F, Guo R, Cao H, 2020. Information coefficient as a performance measure of stock selection models. https://arxiv.org/abs/2010.08601

[103]Zhang L, Aggarwal C, Qi GJ, 2017. Stock price prediction via discovering multi-frequency trading patterns. Proc 23^rd ACM SIGKDD Int Conf on Knowledge Discovery and Data Mining, p.2141-2149.

[104]Zhang Y, Zhao P, Wu Q, et al., 2022. Cost-sensitive portfolio selection via deep reinforcement learning. IEEE Trans Knowl Data Eng, 34(1):236-248.

[105]Zhang Z, Zohren S, Roberts S, 2019. DeepLOB: deep convolutional neural networks for limit order books. IEEE Trans Signal Process, 67(11):3001-3012.

[106]Zhang Z, Lim B, Zohren S, 2021. Deep learning for market by order data. Appl Math Fin, 28(1):79-95.

[107]Zhao J, Zhang C, Qin M, et al., 2025. QuantFactor REINFORCE: mining steady formulaic alpha factors with variance-bounded REINFORCE. IEEE Trans Signal Process, 73:2448-2463.

[108]Zhong L, Wu J, Li Q, et al., 2023. A comprehensive survey on automatic knowledge graph construction. ACM Comput Surv, 56(4):1-62.

[109]Zhou C, Li Q, Li C, et al., 2024. A comprehensive survey on pretrained foundation models: a history from BERT to ChatGPT. Int J Mach Learn Cybern.