Universal features of price formation in financial markets: perspectives from deep learning

By Justin A. Sirignano, Rama Cont. 2018

360 citations recorded by the source corpus.

Using a large-scale Deep Learning approach applied to a high-frequency database containing billions of market quotes and transactions for US equities, we uncover nonparametric evidence for the existence of a universal and stationary relation between order flow history and the direction of price moves. The universal price formation model exhibits a remarkably stable out-of-sample accuracy across a wide range of stocks and time periods. Interestingly, these results also hold for stocks which are not part of the training sample, showing that the relations captured by the model are universal and not asset-specific. The universal model—trained on data from all stocks—outperforms asset-specific models trained on time series of any given stock. This weighs in favor of pooling together financial data from various stocks, rather than designing asset- or sector-specific models, as is currently commonly done. Standard data normalizations based on volatility, price level or average spread, or partitioning the training data into sectors or categories such as large/small tick stocks, do not improve training results. On the other hand, inclusion of price and order flow history over many past observations improves forecast accuracy, indicating that there is path-dependence in price dynamics.

Subfields: Economics and Econometrics, Finance, Management Science and Operations Research, Economics & Operations Research, Time-Series & Forecasting

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Economics and Econometrics

In Finance

In Management Science and Operations Research

In Economics & Operations Research

In Time-Series & Forecasting

Citations in the corpus

Financial Time Series Forecasting with Deep Learning : A Systematic Literature Review: 2005-2019 (2019, methodology): “The authors of [217] proposed a novel method that used limit order book flow and history information for the determination of the stock movements using LSTM.”
Enhancing Time-Series Momentum Strategies Using Deep Neural Networks (2019, methodology): “…long been used for financial time-series prediction, with recent deep learning applications studying mid-price prediction using daily data (Ghoshal and Roberts 2018) or using limit order book data in a high-frequency trading setting (Sirignano and Cont 2018; Zhang, Zohren, and Roberts 2018, 2019).”
Deep Learning for Financial Applications : A Survey (2020, methodology): “[55] High-frequency record of all orders 2014-2017 Price data, record of all orders, trans-”
Mean Field Analysis of Neural Networks: A Law of Large Numbers (2018, methodology): “In finance, neural networks have been used to model loan default and prepayment risk [43] and to model high frequency financial data [44].”
Enhancing the momentum strategy through deep regression (2019, methodology): “Deep learning has also been utilized for other domains in ﬁnance such as for high frequency trading applications in Dixon et al. (2018) and Sirignano and Cont (2018), for options pricing in Berner et al. (2018) and for mortgage risk assessments in Sirignano et al. (2018).”
Cryptocurrency trading: a comprehensive survey (2020, background): “[197] that universal models have better performance than currency-pair specific models for cryptocurrency markets.”
A Survey of Forex and Stock Price Prediction Using Deep Learning (2021, background): “It has been proven that the fluctuation in stock and Forex price could be predicted [5].”
Mean Field Analysis of Deep Neural Networks (2019, background): “Neural networks also find increasing more applications in engineering, robotics, medicine, and finance (see [21], [22], [43], [15], [32], [1], [36], [37], [38], and [39]).”
DeepLOB: Deep Convolutional Neural Networks for Limit Order Books (2018, methodology): “Unlike CNNs which are less widely applied in ﬁnancial markets, the LSTM has been popular in recent years, [42, 28, 43, 44, 45, 46, 47, 20] all utilising LSTMs to analyse ﬁnancial data.”