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

Justin Sirignano, Rama Cont · 2019 · Quantitative Finance

AdvancedHFTEmpiricalFree to read

One deep model trained on hundreds of stocks predicts price direction better than stock-specific models: price formation looks universal.

Read the paper ↗Free version ↗

At a glance

Level: Advanced: Stochastic control, point processes, heavy econometrics or deep learning
Horizon: HFT
Data needed: L2 depth (every update) Set what data you have
Themes: Machine learning Order-flow alpha
Assets: Equities
Market: US

Your notes

Report a problem with this entry

Step 4 of 6 in

Deep learning on the order book (with a reality check)

← DeepLOB: Deep Convolutional Neural Networks for Limit Order BooksDeep order flow imbalance: Extracting alpha at multiple horizons from the limit order book →

Cite

@article{sirignano2019universal,
  author = {Sirignano, Justin and Cont, Rama},
  title = {{Universal features of price formation in financial markets: perspectives from deep learning}},
  journal = {Quantitative Finance},
  year = {2019},
  doi = {10.1080/14697688.2019.1622295}
}

Sirignano, J., & Cont, R. (2019). Universal features of price formation in financial markets: perspectives from deep learning. Quantitative Finance. https://doi.org/10.1080/14697688.2019.1622295