TY - JOUR
T1 - Universal features of price formation in financial markets
T2 - perspectives from deep learning
AU - Sirignano, Justin
AU - Cont, Rama
N1 - Publisher Copyright:
© 2019, © 2019 Informa UK Limited, trading as Taylor & Francis Group.
PY - 2019
Y1 - 2019
N2 - 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.
AB - 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.
KW - Deep learning
KW - Financial econometrics
KW - High-frequency data
KW - Intraday data
KW - Limit order book
KW - Machine learning
KW - Market microstructure
KW - Price formation
UR - http://www.scopus.com/inward/record.url?scp=85068677237&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85068677237&partnerID=8YFLogxK
U2 - 10.1080/14697688.2019.1622295
DO - 10.1080/14697688.2019.1622295
M3 - Article
AN - SCOPUS:85068677237
SN - 1469-7688
VL - 19
SP - 1449
EP - 1459
JO - Quantitative Finance
JF - Quantitative Finance
IS - 9
ER -