Network Ensemble Model for Trend Analysis of Limit Order Books
LÜ Xuerui1, ZHANG Li1,2
1. School of Computer Science and Technology, Soochow University, Suzhou 215006 2. Joint International Research Laboratory of Machine Learning and Neuromorphic Computing, Soochow University, Suzhou 215006
Abstract To analyze the trend of limit order books(LOBs) better, a network ensemble model for trend analysis of LOBs(NEM-LOB) is proposed. Two long short-term memory(LSTM) sub-models and one convolutional neural network sub-model are integrated in NEM-LOB. One LSTM sub-model captures the global temporal dependence through the distribution information of LOBs. The other LSTM sub-model captures the global dynamics through the dynamic information of LOBs and order streams. The local features are extracted through the factual information of LOBs. Finally, three sub-models are combined to extract features to obtain prediction results. Experiments on FI-2010 dataset show that NEM-LOB makes a better trend analysis for LOBs by combining order streams.
Fund:Natural Science Foundation of Jiangsu Higher Education Institutions(No.19KJA550002), Six Talent Peak Project of Jiangsu Province(No.XYDXX-054), Project Funded by the Priority Academic Program Development of Jiangsu Higher Education Institutions
Corresponding Authors:
ZHANG Li, Ph.D., professor. Her research interests include machine learning, pattern recognition and image processing.
About author:: LÜ Xuerui, master student. Her research interests include machine learning, deep learning and financial time series analysis.
[1] HU G S, HU Y X, YANG K, et al. Deep Stock Representation Learning: From Candlestick Charts to Investment Decisions // Proc of the International Conference on Acoustics, Speech, and Signal Processing. Washington, USA: IEEE, 2018: 2706-2710. [2] CONT R. Statistical Modeling of High-Frequency Financial Data. IEEE Signal Processing Magazine, 2011, 28(5): 16-25. [3] WEI L J, SHI L. Investor Sentiment in an Artificial Limit Order Market[C/OL]. [2021-03-16]. https://downloads.hindawi.com/journals/complexity/2020/8581793.pdf. [4] 杨宝臣,郭 灿,王 硕.基于限价指令簿信息的流动性冲击影响.系统工程, 2017, 35(3): 21-28. (YANG B C, GUO C, WANG S. Order Flows′ Liquidity Shock Based on the Order-Driven Market. Systems Engineering, 2017, 35(3): 21-28.) [5] 张传海,汪璐欹,彭 哲.限价指令簿信息对价格动态的非对称效应——来自我国A股市场的经验研究.武汉金融, 2020(3): 18-27. (ZHANG C H, WANG L Q, PENG Z. Asymmetric Effect of Limit Order Book Information on Price Dynamics: An Empirical Study from My Country′s A-Shares Market. Wuhan Finance, 2020(3): 18-27.) [6] TRAN D T, MAGRIS M, KANNIAINEN J, et al. Tensor Representation in High-Frequency Financial Data for Price Change Prediction // Proc of the IEEE Symposium Series on Computational Intelligence. Washington, USA: IEEE, 2017. DOI: 10.1109/SSCI.2017.8280812. [7] NTAKARIS A, MAGRISV M, KANNIAINEN J, et al. Benchmark Dataset for Mid-price Forecasting of Limit Order Book Data with Machine Learning Methods[C/OL]. [2021-03-16].https://arxiv.org/pdf/1705.03233v4.pdf. [8] KERCHEVAL A N, ZHANG Y. Modelling High-Frequency Limit Order Book Dynamics with Support Vector Machines. Quantitative Finance, 2015, 15(8): 1315-1329. [9] NTAKARIS A, KANNIAINEN J, GABBOUJ M, et al. Mid-price Prediction Based on Machine Learning Methods with Technical and Quantitative Indicators[C/OL]. [2021-03-16].https://arxiv.org/pdf/1907.09452v1.pdf. [10] TSANTEKIDIS A, PASSALIS N, TEFAS A, et al. Forecasting Stock Prices from the Limit Order Book Using Convolutional Neural Networks // Proc of the 19th IEEE Conference on Business Informatics. Washington, USA: IEEE, 2017: 7-12. [11] NIÑO J, ARÉVALO A, LEÓN D, et al. Price Prediction with CNN and Limit Order Book Data // Proc of the Workshop on Engineering Applications. Berlin, Germany: Springer, 2018: 124-135. [12] PASSALIS N, TEFAS A, KANNIAINEN J, et al. Adaptive Normalization for Forecasting Limit Order Book Data Using Convolutional Neural Networks // Proc of the IEEE International Confe-rence on Acoustics, Speech, and Signal Processing. Washington, USA: IEEE, 2020: 1713-1717. [13] MÄKINEN Y, KANNIAINEN J, GABBOUJ M, et al. Forecasting Jump Arrivals in Stock Prices: New Attention-Based Network Architecture Using Limit Order Book Data. Quantitative Finance, 2019, 19(12): 2033-2050. [14] ZHANG Z H, ZOHREN S, ROBERTS S. DeepLOB: Deep Convolutional Neural Networks for Limit Order Books. IEEE Transactions on Signal Processing, 2019, 67(11): 3001-3012. [15] TRAN D T, IOSIFIDIS A, KANNIAINEN J, et al. Temporal Attention-Augmented Bilinear Network for Financial Time-Series Data Analysis. IEEE Transactions on Neural Networks and Learning Systems, 2018, 30(5): 1407-1418. [16] NAIK N, MOHAN B R. Intraday Stock Prediction Based on Deep Neural Network. National Academy Science Letters, 2020, 43(2): 241-246. [17] CHOI S P, CHAN Y, LAM S, et al. Analyzing the Importance of Broker Identities in the Limit Order Book through Deep Learning. Big Data, 2021,9(2): 89-99. [18] GOULD M D, PORTER M A, WILLIAMS S, et al. Limit Order Books. Quantitative Finance, 2013, 13(11): 1709-1742. [19] LI J Y, WANG X T, LIN Y Y, et al. Generating Realistic Stock Market Order Streams. Proceedings of the AAAI Conference on Artificial Intelligence, 2020, 34(1): 727-734. [20] HOCHREITER S, SCHMIDHUBER J. Long Short-Term Memory. Neural Computation, 1997, 9(8): 1735-1780. [21] ZHAO R, WANG K, SU H, et al. Bayesian Graph Convolution LSTM for Skeleton Based Action Recognition // Proc of the IEEE/CVF International Conference on Computer Vision. Washington, USA: IEEE, 2019: 6881-6891. [22] TRAN D T, GABBOUJ M, IOSIFIDIS A. Multilinear Class-Speci-fic Discriminant Analysis. Pattern Recognition Letters, 2017, 100: 131-136. [23] PASSALIS N, TSANTEKIDIS A, TEFAS A, et al. Time-Series Classification Using Neural Bag-of-Features // Proc of the 25th European Signal Processing Conference. Washington, USA: IEEE, 2017: 301-305. [24] PASSALIS N, TEFAS A, KANNIAINEN J, et al. Deep Temporal Logistic Bag-of-Features for Forecasting High Frequency Limit Order Book Time Series // Proc of the IEEE International Conference on Acoustics, Speech and Signal Processing. Washington, USA: IEEE, 2019: 7545-7549. [25] PASSALIS N, TEFAS A, KANNIAINEN J, et al. Temporal Logistic Neural Bag-of-Features for Financial Time Series Forecasting Leveraging Limit Order Book Data. Pattern Recognition Letters, 2020, 136: 183-189. [26] TSANTEKIDIS A, PASSALIS N, TEFAS A, et al. Using Deep Learning to Detect Piece Change Indications in Financial Markets // Proc of the 25th European Signal Processing Conference. Washington, USA: IEEE, 2017: 2511-2515. [27] TSANTEKIDIS A, PASSALIS N, TEFAS A, et al. Using Deep Learning for Price Prediction by Exploiting Stationary Limit Order Book Features[C/OL]. [2021-03-16]. https://arxiv.org/pdf/1810.09965.pdf.
2021, Vol. 34 
