Dynamic Portfolio Construction in High-Frequency Markets Using Microstructure-Aware Deep Temporal Models

Article Sidebar

PDF
Published: 2026-03-04
Keywords:
high-frequency trading, portfolio construction, market microstructure, deep temporal models, order flow, risk-adjusted performance

Main Article Content

Xiaoyi Meng
University of California, Irvine, 92618, USA

Abstract

High-frequency financial markets exhibit rapidly changing liquidity and strong nonstationarity, which reduce the effectiveness of portfolio methods based on static return and covariance estimates. This study presents a microstructure-aware framework for dynamic portfolio allocation that maps high-frequency market states directly to portfolio weights. The model uses order-flow intensity, depth imbalance, and volume asymmetry as key inputs and combines recurrent and convolutional temporal components to capture both persistent dependencies and short-lived microstructure shocks. Portfolio weights are determined under a joint return–risk objective. The framework is evaluated on ten years of high-frequency U.S. equity data. Results show higher risk-adjusted performance than mean–variance and risk-parity benchmarks, with a 14%–22% improvement in Sharpe ratio and more stable drawdown behavior. These findings suggest that integrating market microstructure information into temporal allocation models can improve exposure management in intraday portfolio optimization.

Article Details

Issue

Vol. 2 No. 2 (2026)

Section

Articles

How to Cite

Dynamic Portfolio Construction in High-Frequency Markets Using Microstructure-Aware Deep Temporal Models. (2026). Journal of Sustainability, Policy, and Practice, 2(2), 1-6. https://schoalrx.com/index.php/jspp/article/view/91

References

1. P. N. Kolm, J. Turiel, and N. Westray, "Deep order flow imbalance: Extracting alpha at multiple horizons from the limit order book," Mathematical Finance, vol. 33, no. 4, pp. 1044-1081, 2023. doi: 10.1111/mafi.12413

2. L. Zhang, and L. Hua, "Major issues in high-frequency financial data analysis: A survey of solutions," Mathematics, vol. 13, no. 3, p. 347, 2025. doi: 10.3390/math13030347

3. A. Briola, S. Bartolucci, and T. Aste, "Deep limit order book forecasting: a microstructural guide," Quantitative Finance, vol. 25, no. 7, pp. 1101-1131, 2025. doi: 10.1080/14697688.2025.2522911

4. Z. Eisler, J. P. Bouchaud, and J. Kockelkoren, "The price impact of order book events: market orders, limit orders and cancellations," Quantitative Finance, vol. 12, no. 9, pp. 1395-1419, 2012.

5. N. Hautsch, L. M. Kyj, and P. Malec, "The merit of high-frequency data in portfolio allocation (No. 2011-059)," SFB 649 Discussion Paper, 2011.

6. S. Giantsidi, and T. Claudia, "Deep Learning for Financial Forecasting: A Review of Recent Advancements," Available at SSRN 5263710. doi: 10.2139/ssrn.5263710

7. P. Lara-Benítez, M. Carranza-García, J. M. Luna-Romera, and J. C. Riquelme, "Temporal convolutional networks applied to energy-related time series forecasting," applied sciences, vol. 10, no. 7, p. 2322, 2020. doi: 10.3390/app10072322

8. Q. Zhao, H. Li, Q. Zhang, and Y. Wang, "A Study on Metal Futures Price Prediction Based on Piecewise Cubic Bézier Filtering for TCN," Applied Sciences, vol. 15, no. 17, p. 9792, 2025. doi: 10.3390/app15179792

9. Y. Guo, and X. Chen, "Forecasting the mid-price movements with high-frequency lob: A dual-stage temporal attention-based deep learning architecture," Arabian Journal for Science and Engineering, vol. 48, no. 8, pp. 9597-9618, 2023. doi: 10.1007/s13369-022-07197-3

10. K. Kubo, and K. Nakagawa, "Portfolio optimization using deep learning with risk aversion utility function," Finance Research Letters, vol. 74, p. 106761, 2025. doi: 10.1016/j.frl.2025.106761

11. J. Zou, F. Wang, and Y. Wu, "Large portfolio allocation using high-frequency financial data," 2018. doi: 10.4310/sii.2018.v11.n1.a12

12. Z. Zhang, S. Zohren, and S. Roberts, "Deep learning for portfolio optimization," arXiv preprint arXiv:2005.13665, 2020. doi: 10.3905/jfds.2020.1.042

13. O. Ali, "Deep Learning for Portfolio Optimization: An AI-Driven Approach to Risk-Adjusted Returns," Available at SSRN 5510579.

14. X. Y. Liu, Z. Xia, J. Rui, J. Gao, H. Yang, M. Zhu, and J. Guo, "Finrl-meta: Market environments and benchmarks for data-driven financial reinforcement learning," Advances in Neural Information Processing Systems, vol. 35, pp. 1835-1849, 2022.

15. N. Passalis, A. Tefas, J. Kanniainen, M. Gabbouj, and A. Iosifidis, "Temporal logistic neural bag-of-features for financial time series forecasting leveraging limit order book data," Pattern Recognition Letters, vol. 136, pp. 183-189, 2020. doi: 10.1016/j.patrec.2020.06.006

16. S. Kim, H. Kwon, J. Yoon, and H. Y. Kim, "Dynamic factor-informed reinforcement learning for enhancing portfolio optimization," Financial Innovation, vol. 12, no. 1, p. 3, 2026. doi: 10.1186/s40854-025-00803-x

17. R. Paki, and H. Abbasimehr, "PB-TABL: Task Incremental Learning Strategy via Applying Piggyback Architecture on Temporal Attention-Augmented Bilinear Networks for Financial Time-Series Classification," IEEE Transactions on Big Data, 2025. doi: 10.1109/tbdata.2025.3598730

18. Z. Wang, B. Huang, S. Tu, K. Zhang, and L. Xu, "Deeptrader: a deep reinforcement learning approach for risk-return balanced portfolio management with market conditions embedding," In Proceedings of the AAAI conference on artificial intelligence, May, 2021, pp. 643-650. doi: 10.1609/aaai.v35i1.16144