Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
157 tokens/sec
GPT-4o
8 tokens/sec
Gemini 2.5 Pro Pro
46 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

MarS: a Financial Market Simulation Engine Powered by Generative Foundation Model (2409.07486v2)

Published 4 Sep 2024 in q-fin.CP, cs.AI, cs.CE, cs.LG, and q-fin.TR

Abstract: Generative models aim to simulate realistic effects of various actions across different contexts, from text generation to visual effects. Despite significant efforts to build real-world simulators, the application of generative models to virtual worlds, like financial markets, remains under-explored. In financial markets, generative models can simulate complex market effects of participants with various behaviors, enabling interaction under different market conditions, and training strategies without financial risk. This simulation relies on the finest structured data in financial market like orders thus building the finest realistic simulation. We propose Large Market Model (LMM), an order-level generative foundation model, for financial market simulation, akin to LLMing in the digital world. Our financial Market Simulation engine (MarS), powered by LMM, addresses the domain-specific need for realistic, interactive and controllable order generation. Key observations include LMM's strong scalability across data size and model complexity, and MarS's robust and practicable realism in controlled generation with market impact. We showcase MarS as a forecast tool, detection system, analysis platform, and agent training environment, thus demonstrating MarS's "paradigm shift" potential for a variety of financial applications. We release the code of MarS at https://github.com/microsoft/MarS/.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (64)
  1. Phi-3 technical report: A highly capable language model locally on your phone. arXiv preprint arXiv:2404.14219, 2024.
  2. Gpt-4 technical report. arXiv preprint arXiv:2303.08774, 2023.
  3. Direct estimation of equity market impact. 2005.
  4. Abides-gym: gym environments for multi-agent discrete event simulation and application to financial markets. In Proceedings of the Second ACM International Conference on AI in Finance, pp.  1–9, 2021.
  5. Market impacts and the life cycle of investors orders, 2014. URL https://arxiv.org/abs/1412.0217.
  6. Market impacts and the life cycle of investors orders. Market Microstructure and Liquidity, 1(02):1550009, 2015.
  7. Fintral: A family of gpt-4 level multimodal financial large language models. arXiv preprint arXiv:2402.10986, 2024.
  8. On the opportunities and risks of foundation models. arXiv preprint arXiv:2108.07258, 2021.
  9. Video generation models as world simulators. 2024. URL https://openai.com/research/video-generation-models-as-world-simulators.
  10. Tom B Brown. Language models are few-shot learners. arXiv preprint arXiv:2005.14165, 2020.
  11. Abides: Towards high-fidelity multi-agent market simulation. In Proceedings of the 2020 ACM SIGSIM Conference on Principles of Advanced Discrete Simulation, pp.  11–22, 2020.
  12. The impact of heterogeneous trading rules on the limit order book and order flows. Journal of Economic Dynamics and Control, 33(3):525–537, 2009.
  13. Towards realistic market simulations: a generative adversarial networks approach. In Proceedings of the Second ACM International Conference on AI in Finance, pp.  1–9, 2021.
  14. Learning to simulate realistic limit order book markets from data as a world agent. In Proceedings of the third acm international conference on ai in finance, pp.  428–436, 2022.
  15. Conditional generators for limit order book environments: Explainability, challenges, and robustness. In Proceedings of the Fourth ACM International Conference on AI in Finance, pp.  27–35, 2023.
  16. Optimal execution with non-linear transient market impact. Quantitative Finance, 17(1):41–54, 2017.
  17. A fully consistent, minimal model for non-linear market impact. Quantitative finance, 15(7):1109–1121, 2015a.
  18. A fully consistent, minimal model for non-linear market impact, 2015b. URL https://arxiv.org/abs/1412.0141.
  19. Learning universal policies via text-guided video generation. In A. Oh, T. Naumann, A. Globerson, K. Saenko, M. Hardt, and S. Levine (eds.), Advances in Neural Information Processing Systems, volume 36, pp.  9156–9172. Curran Associates, Inc., 2023. URL https://proceedings.neurips.cc/paper_files/paper/2023/file/1d5b9233ad716a43be5c0d3023cb82d0-Paper-Conference.pdf.
  20. The llama 3 herd of models. arXiv preprint arXiv:2407.21783, 2024.
  21. Jim Gatheral. No-dynamic-arbitrage and market impact. Quantitative finance, 10(7):749–759, 2010.
  22. Exponential resilience and decay of market impact. Econophysics of Order-driven Markets: Proceedings of Econophys-Kolkata V, pp.  225–236, 2011.
  23. Transient linear price impact and fredholm integral equations. Mathematical Finance: An International Journal of Mathematics, Statistics and Financial Economics, 22(3):445–474, 2012.
  24. Generative adversarial networks. Communications of the ACM, 63(11):139–144, 2020.
  25. Limit order books. Quantitative Finance, 13(11):1709–1742, 2013.
  26. Stock trend prediction with multi-granularity data: A contrastive learning approach with adaptive fusion. In Proceedings of the 30th ACM International Conference on Information & Knowledge Management, pp.  700–709, 2021.
  27. Lawyer llama technical report. arXiv preprint arXiv:2305.15062, 2023.
  28. A generative model of a limit order book using recurrent neural networks. Quantitative Finance, 23(6):931–958, 2023.
  29. Scaling laws for neural language models. arXiv preprint arXiv:2001.08361, 2020.
  30. Hats: A hierarchical graph attention network for stock movement prediction. arXiv preprint arXiv:1908.07999, 2019.
  31. Generating realistic stock market order streams. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 34, pp.  727–734, 2020.
  32. Synthetic data generation with large language models for text classification: Potential and limitations. arXiv preprint arXiv:2310.07849, 2023.
  33. Master curve for price-impact function. Nature, 421(6919):129–130, 2003. doi: 10.1038/421129a. URL https://doi.org/10.1038/421129a.
  34. Visual instruction tuning, 2023a.
  35. Fingpt: Democratizing internet-scale data for financial large language models. arXiv preprint arXiv:2307.10485, 2023b.
  36. Sora: A review on background, technology, limitations, and opportunities of large vision models. arXiv preprint arXiv:2402.17177, 2024.
  37. Lisa K Meulbroek. An empirical analysis of illegal insider trading. The Journal of Finance, 47(5):1661–1699, 1992.
  38. Augmented language models: a survey. arXiv preprint arXiv:2302.07842, 2023.
  39. Foundation models for generalist medical artificial intelligence. Nature, 616(7956):259–265, 2023.
  40. Market impact and trading profile of hidden orders in stock markets. Physical Review E, 80(6), December 2009. ISSN 1550-2376. doi: 10.1103/physreve.80.066102. URL http://dx.doi.org/10.1103/PhysRevE.80.066102.
  41. 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. In Proceedings of the Fourth ACM International Conference on AI in Finance, pp.  91–99, 2023.
  42. Benchmark dataset for mid‐price forecasting of limit order book data with machine learning methods. Journal of Forecasting, 37(8):852–866, August 2018. ISSN 1099-131X. doi: 10.1002/for.2543. URL http://dx.doi.org/10.1002/for.2543.
  43. Tālis J Putniņš. Market manipulation: A survey. Journal of economic surveys, 26(5):952–967, 2012.
  44. Deep state space models for time series forecasting. Advances in neural information processing systems, 31, 2018.
  45. High-resolution image synthesis with latent diffusion models, 2021.
  46. Study of stylized facts in stock market data, 2023. URL https://arxiv.org/abs/2310.00753.
  47. Deep calibration of market simulations using neural density estimators and embedding networks, 2023. URL https://arxiv.org/abs/2311.11913.
  48. Reinforcement Learning: An Introduction. MIT Press, second edition, 2018.
  49. Modeling financial time-series with generative adversarial networks. Physica A: Statistical Mechanics and its Applications, 527:121261, 2019.
  50. Get real: Realism metrics for robust limit order book market simulations. In Proceedings of the First ACM International Conference on AI in Finance, pp.  1–8, 2020.
  51. Pedram Babaei William Todt, Ramtin Babaei. Fin-llama: Efficient finetuning of quantized llms for finance. https://github.com/Bavest/fin-llama, 2023.
  52. Bloomberggpt: A large language model for finance. arXiv preprint arXiv:2303.17564, 2023.
  53. Pixiu: A comprehensive benchmark, instruction dataset and large language model for finance. Advances in Neural Information Processing Systems, 36, 2024a.
  54. Open-finllms: Open multimodal large language models for financial applications. arXiv preprint arXiv:2408.11878, 2024b.
  55. Learning interactive real-world simulators. arXiv preprint arXiv:2310.06114, 2023a.
  56. Video as the new language for real-world decision making. arXiv preprint arXiv:2402.17139, 2024.
  57. Investlm: A large language model for investment using financial domain instruction tuning. arXiv preprint arXiv:2309.13064, 2023b.
  58. Beyond the square root: Evidence for logarithmic dependence of market impact on size and participation rate. Market Microstructure and Liquidity, 1(02):1550004, 2015.
  59. Scaling vision transformers. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp.  12104–12113, 2022.
  60. Instruct-fingpt: Financial sentiment analysis by instruction tuning of general-purpose large language models. arXiv preprint arXiv:2306.12659, 2023.
  61. Reinforcement learning with maskable stock representation for portfolio management in customizable stock pools. In Proceedings of the ACM on Web Conference 2024, pp.  187–198, 2024.
  62. Xuanyuan 2.0: A large chinese financial chat model with hundreds of billions parameters. In Proceedings of the 32nd ACM international conference on information and knowledge management, pp.  4435–4439, 2023.
  63. Deeplob: Deep convolutional neural networks for limit order books. IEEE Transactions on Signal Processing, 67(11):3001–3012, June 2019. ISSN 1941-0476. doi: 10.1109/tsp.2019.2907260. URL http://dx.doi.org/10.1109/TSP.2019.2907260.
  64. Is sora a world simulator? a comprehensive survey on general world models and beyond. arXiv preprint arXiv:2405.03520, 2024.

Summary

We haven't generated a summary for this paper yet.