The Limit Order Book Recreation Model (LOBRM): An Extended Analysis

The limit order book (LOB) depicts the fine-grained demand and supply relationship for financial assets and is widely used in market microstructure studies. Nevertheless, the availability and high cost of LOB data restrict its wider application. The LOB recreation model (LOBRM) was recently proposed to bridge this gap by synthesizing the LOB from trades and quotes (TAQ) data. However, in the original LOBRM study, there were two limitations: (1) experiments were conducted on a relatively small dataset containing only one day of LOB data; and (2) the training and testing were performed in a non-chronological fashion, which essentially re-frames the task as interpolation and potentially introduces lookahead bias. In this study, we extend the research on LOBRM and further validate its use in real-world application scenarios. We first advance the workflow of LOBRM by (1) adding a time-weighted z-score standardization for the LOB and (2) substituting the ordinary differential equation kernel with an exponential decay kernel to lower computation complexity. Experiments are conducted on the extended LOBSTER dataset in a chronological fashion, as it would be used in a real-world application. We find that (1) LOBRM with decay kernel is superior to traditional non-linear models, and module ensembling is effective; (2) prediction accuracy is negatively related to the volatility of order volumes resting in the LOB; (3) the proposed sparse encoding method for TAQ exhibits good generalization ability and can facilitate manifold tasks; and (4) the influence of stochastic drift on prediction accuracy can be alleviated by increasing historical samples.

[1]  David Duvenaud,et al.  Latent Ordinary Differential Equations for Irregularly-Sampled Time Series , 2019, NeurIPS.

[2]  Michael Kearns,et al.  Reinforcement learning for optimized trade execution , 2006, ICML.

[3]  I. Florescu,et al.  Extracting Information From the Limit Order Book: New Measures to Evaluate Equity Data Flow , 2019, High Frequency.

[4]  Ash Booth,et al.  High frequency trading strategies, market fragility and price spikes: an agent based model perspective , 2018, Annals of Operations Research.

[5]  Alexandros Iosifidis,et al.  Temporal Logistic Neural Bag-of-Features for Financial Time series Forecasting leveraging Limit Order Book Data , 2019, Pattern Recognit. Lett..

[6]  Justin A. Sirignano,et al.  Universal features of price formation in financial markets: perspectives from deep learning , 2018, Machine Learning and AI in Finance.

[7]  Rama Cont,et al.  Universal features of price formation in financial markets: perspectives from deep learning , 2019, Quantitative Finance.

[8]  Ulrich Horst,et al.  A Weak Law of Large Numbers for a Limit Order Book Model with Fully State Dependent Order Dynamics , 2015, SIAM J. Financial Math..

[9]  Pankaj Kumar.  Deep Reinforcement Learning for Market Making , 2020, AAMAS.

[10]  Alexandros Iosifidis,et al.  Using Deep Learning for price prediction by exploiting stationary limit order book features , 2018, Appl. Soft Comput..

[11]  Rama Cont,et al.  Price Dynamics in a Markovian Limit Order Market , 2011, SIAM J. Financial Math..

[12]  Zijian Shi,et al.  The LOB Recreation Model: Predicting the Limit Order Book from TAQ History Using an Ordinary Differential Equation Recurrent Neural Network , 2020, AAAI.

[13]  Jonas Gloeckner,et al.  The Double Auction Market Institutions Theories And Evidence , 2016 .

[14]  D. Friedman.  The Double Auction Market Institution: A Survey , 2018 .

[15]  Michael P. Wellman,et al.  Generating Realistic Stock Market Order Streams , 2020, AAAI.

[16]  R. Palmer,et al.  Asset Pricing Under Endogenous Expectations in an Artificial Stock Market , 1996 .

[17]  Huyen Pham,et al.  Algorithmic trading in a microstructural limit order book model , 2017, Quantitative Finance.

[18]  Mathias Lechner,et al.  Learning Long-Term Dependencies in Irregularly-Sampled Time Series , 2020, NeurIPS.

[19]  Christine A. Parlour.  Price Dynamics in Limit Order Markets , 1998 .

[20]  J. Blanchet,et al.  Unraveling Limit Order Books Using Just Bid/Ask Prices , 2017 .

[21]  Nigel P. Smart,et al.  Multi-Party Computation Mechanism for Anonymous Equity Block Trading: A Secure Implementation of Turquoise Plato Uncross , 2021, IACR Cryptol. ePrint Arch..

[22]  Yan Liu,et al.  Recurrent Neural Networks for Multivariate Time Series with Missing Values , 2016, Scientific Reports.

[23]  Stefan Zohren,et al.  Fast Agent-Based Simulation Framework of Limit Order Books with Applications to Pro-Rata Markets and the Study of Latency Effects , 2020, ArXiv.

[24]  Yoshua Bengio,et al.  Learning Phrase Representations using RNN Encoder–Decoder for Statistical Machine Translation , 2014, EMNLP.

[25]  Stefan Zohren,et al.  DeepLOB: Deep Convolutional Neural Networks for Limit Order Books , 2018, IEEE Transactions on Signal Processing.

[26]  Nigel P. Smart,et al.  MPC Joins The Dark Side , 2019, IACR Cryptol. ePrint Arch..

[27]  Aymen Jedidi,et al.  Long-Time Behavior of a Hawkes Process-Based Limit Order Book , 2015, SIAM J. Financial Math..

[28]  M. Gabbouj,et al.  Forecasting jump arrivals in stock prices: new attention-based network architecture using limit order book data , 2018, Quantitative Finance.