Backtester Java

A modular backtesting engine for financial trading strategies, written in Java. This is still work in progress, but it is already functional.

Overview

This project allows you to test and compare trading strategies on historical data. The system is modular, so stock universes, strategies and signals can easily be switched. The system takes CSV-files on Date-OHLCV format. They can easily be downloaded from Yahoo Finance using import_data.py. The script now accepts tickers and date ranges on the command line for quick test setup

Examples

Features

• Strategy Framework: Implement and plug in custom strategies using existing signals if desired
• Portfolio Simulation: Simulate portfolio value, positions, and trades over time
• Multiple Built-in Strategies: Includes EMA, Momentum, Stop Loss, and Buy & Hold strategies
• Performance Analytics: Summarizes returns, cash, positions, and more
• Visualization: Plots portfolio value over time using XChart
• Logging: Detailed logging for debugging and analysis

Project Structure

• src/main/java/app/ – Orchestration and entry point (App.java, BacktestOrchestrator.java)
• src/main/java/strategies/ – Strategy implementations (e.g., EMAStrategy, MomentumStrategy)
• src/main/java/accounts/ – Portfolio, Order, and related classes
• src/main/java/io/ – Logging utilities
• src/main/java/engine/ – Market data and simulation engine

Getting Started

Prerequisites

• Java 21
• Maven

Build and Run

1. Fetch example data using the helper script:

   python import_data.py AAPL MSFT SPY --start 2000-01-01 --end 2010-01-01

2. Compile and run the tests:

   mvn test

3. Launch the backtester (charts are skipped automatically when no display is available):

   mvn exec:java

   The orchestrator will write each strategy's performance to CSV files under the results/ directory.

4. Visualize the results:

   python visualize_results.py results

Usage Example

Modify App.java to configure which strategies to test and with what parameters. Example strategies include:

• Buy & Hold
• EMA Crossover
• Momentum
• Stop Loss

After running, the orchestrator will print portfolio summaries and plot performance charts.

Extending

To add your own strategy:

1. Create a new class in strategies/ extending Strategy.
2. Implement the onBars method with your trading logic.
3. Register your strategy in App.java using BacktestOrchestrator.StrategyConfig.

LLM Usage

This project has been created with the help of ChatGPT-4.1, especially for writing printing and visualization logic, as well as implementing several of the strategies.