midastrader 1.0.1

A robust backtesting and live trading engine designed for seamless strategy development and deployment. It supports user-defined strategies, multi-threaded execution, and integrations with brokers and data sources.

MidasTrader

Overview

MidasTrader is a robust trading system designed for seamless transitions between backtesting and live trading without requiring changes to user code. It integrates a flexible architecture combining a data engine, execution engine, and core components that streamline the strategy development process. The system is built with a multi-threaded design where each component communicates via a shared message bus.

Key Components

1. Core Engine:

   • Central to the system, the Core Engine includes:
     • Order Book: Tracks market depth and price movements.
     • Portfolio Server: Manages and tracks portfolio allocations and positions.
     • Performance Tracking: Calculates and monitors key trading metrics.
     • Order Management System: Handles order placement, modifications, and cancellations.
     • Base Strategy: A foundation for user-defined strategies.

2. Data Engine:

   • Connects to user-defined data sources:
     • Midas Server: Access historical data via the Midas ecosystem.
     • Binary Data Files: Handles local files encoded with the Midas Binary Encoding Library.
     • External Sources: Currently supports Databento, with more integrations planned.

3. Execution Engine:

   • Facilitates live trading by connecting to brokers:
     • Currently supports Interactive Brokers.
     • Users can configure broker details in the config.toml file.

Installation

1. Clone the repository:

   git clone https://github.com/midassystems/midastrader.git
   cd midastrader
   

2. Install dependencies:

   pip install -r requirements.txt
   

3. Install the package:

   python setup.py install
   

Configuration

Configuration File (config.toml)

Define system parameters, including data sources, execution settings, and strategy configuration.

• Example : config.toml

Strategy File (logic.py)

Strategies are implemented by extending the BaseStrategy class. Define your custom logic in Python.

• Example : logic.py

Usage

CLI Mode

Run the system using the following commands:

# Backtest Mode
midas path/to/config.toml backtest

# Live Mode
midas path/to/config.toml live

Application Mode

Alternatively, you can use the system programmatically in your application:

from midas.cli import run

# Backtest Mode
run("path/to/config.toml", "backtest")

# Live Mode
run("path/to/config.toml", "live")

Supported Features

	Data Vendors
✅	Databento

	Brokers
✅	Interactive Brokers

Future Plans

• Add more data sources.
• Integrate additional brokers.

Contributing

Contributions are welcome! Feel free to open an issue or submit a pull request with suggestions or improvements.

License

This project is licensed under the Apache 2.0 License. See the LICENSE file for details.