Skip to content

hudson-and-thames/arbitrage_research


Welcome to Arbitrage Research

Welcome to Arbitrage Research, your gateway to unlocking the potential of quantitative strategy development. Within this repository, you'll find a collection of notebooks showcasing the powerful strategies and tools available in the arbitragelab library, designed to construct mean-reverting portfolios.

Originally conceived as a research endeavor within Hudson and Thames, a pioneering quantitative research firm dedicated to implementing advanced algorithms drawn from leading academic journals and research papers, arbitragelab has long been the secret weapon of research teams within quantitative hedge funds. Now, in a groundbreaking move, the firm has opted to share this powerful resource with the global community, democratizing access for anyone passionate about developing quantitative strategies.

At its core, the library comprises meticulously crafted implementations of statistical arbitrage strategies documented in academic literature, adhering closely to Krauss' taxonomy of Pairs Trading Strategies.

Within this repository, each notebook offers a hands-on exploration of a specific strategy tailored to construct mean-reverting portfolios. The notebooks also provide links and references to papers where the strategy was first proposed. Feel free to check out the links to deepen your understanding of the mathematical and theoretical underpinnings behind each implementation.

Installation

It is recommended to install this project from source using:

git clone https://github.com/hudson-and-thames/arbitrage_research.git
cd arbitrage_research

To install dependencies run

poetry install --no-root

Contributions

As this is a community driven project we welcome any contribution to this project. Please review the contribution guidelines here.

About

Jupyter Notebook examples on how to use the ArbitrageLab - pairs trading - python library.

Resources

License

Code of conduct

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published