Optimal HF trading (HFT & Stochastic Control)

<img src="https://img.shields.io/static/v1?label=Range&message=Academic project&color=007bff"/>  <img src="https://img.shields.io/static/v1?label=Languages&message=Python&color=ff0000"/>

Workflows:

Overview

This project aims at studying a research article involving high-frequency trading & stochastic control applications. One also attempts to reproduce the numerical results shown in this paper.

• Original paper: Optimal high frequency trading with limit and market orders, Fabien GUILBAUD, Huyên PHAM (2011)
  • Link: arxiv (1106.5040) (PDF) (HTML via ar5iv)
• Keywords: Market making, limit order book, inventory risk, point process, stochastic control

Data description

This article's using Level 1 (L1) data from SOGCGEN.PA (intraday data) for only one day: April 18, 2011 between 9:30 and 16:30 in Paris local time (CET, UTC+2).

To reproduce the method introduced in this article, one applies it to L1 (tick data) sample dataset available on TickHistory's website.

Dataset description:

• GOOG: trades and quotes data
• MSFT: trades and quotes data

.
├── tickhistory_sample
│   ├── GOOG_Quote_2022_01_03.txt
│   ├── GOOG_Trade_2022_01_03.txt
│   ├── MSFT_Quote_2022_01_03.txt
│   ├── MSFT_Trade_2022_01_03.txt
│   ├── file_format_details.txt
│   └── tick_history_ticker_coverage.txt
└── tickhistory_sample.zip

Data from:

• https://tickhistory.com/
• https://firstratedata.com/tick-data

Getting started

1. Install required modules:

pip3 install -r requirements.txt

2. Launch a Jupyter instance

jupyter-notebook .

License

Check LICENSE file

• Adrien NAVARRO - Lucas RODRIGUEZ
• Academic works (March - May 2023)