This project(Shannon) is an open-source framework designed to offer efficient backtesting capabilities for multi-factor strategies. The primary objective of this framework is to facilitate researchers and traders in testing, optimizing, and implementing their strategies with ease.

The key to a successful strategy lies in enhancing the signal-to-noise ratio of factors, a concept first introduced by the great foundational figure in information theory, Claude Shannon. Inspired by this concept, the design of this framework thoroughly incorporates various flexible filters to improve the signal-to-noise ratio of factors. Hence, the framework is named "Shannon" as a tribute to this brilliant genius.

Modular Design: The framework is architected to decouple modules like data preparation, strategy formulation, and performance evaluation comprehensively. Users can switch between various modules conveniently, thereby facilitating efficient strategy development.

Strategy-Centric: The focus is on writing the strategy, so users don't have to be distracted by data preparation or performance assessment tasks.

High Flexibility: It combines factor-based stock selection with various filters, offering users a high level of customization. Regardless of the strategy or factors you're based on, you can efficiently implement them within this framework.

After downloading and extracting, switch to the project directory.

pip install -r requirements.txt

1.In config.py, users can configure the data storage path, backtesting parameters, and more.

2.Run preprocess.py for data preprocessing.

3.Execute backtest.py to complete the backtesting and output the strategy evaluation metrics.

4.This framework includes a built-in sample strategy located in the strategies folder. Users can follow its pattern to craft their own strategies.

The data used by this framework includes individual stock trading data,index trading data and financial data. This data needs to be retrieved via an API and then decompressed locally. Users can contact the author to obtain free API access permissions (each type of data can be downloaded once a week).