FuturesBacktestingEngine

Institutional-grade replication of the Société Générale Trend Indicator

Overview

This project is a high-performance futures backtesting engine designed to simulate medium-to-long-term trend-following strategies. Built by a former Hedge Fund Quant and National Lab R&D Engineer, the framework focuses on vectorized performance and modular signal architecture.

It replicates the core logic used in institutional trend-following benchmarks, utilizing moving average crossovers and volatility-adjusted position sizing across a diversified portfolio of 57 synthetic futures markets.

Project Structure

The repository is organized into a modular architecture to separate logic from data artifacts:

• src/: Core Python engine.
  • generator.py: Generates synthetic price data (OHLC) using Brownian Motion.
  • engine.py: Defines market mechanics (Big Point Value, tick sizes).
  • models.py: Contains the 20/120 Simple Moving Average crossover signal logic.
  • main.py: Orchestrates the backtest and performance calculations.
  • visualize_data.py: Debugging tool for data inspection and plotting.
• plots/: Stores the generated strategy equity curve and visualizations.
• data/: (Local only) Stores generated synthetic market history.

Technical Expertise

• Vectorized Backtesting: Optimized for high-speed research using NumPy and Pandas.
• Institutional Risk Metrics: Calculates Sharpe and Sortino ratios to validate strategy robustness.
• Modular Design: Signal logic is decoupled from the execution engine for easy strategy iteration.

Quick Start

1. Initialize the Environment: Ensure you have Python 3.x installed with Pandas, NumPy, and Matplotlib.

2. Generate Market Data: python src/generator.py

3. Run Backtest: python src/main.py

Results

The backtester generates a performance equity curve and risk-adjusted return summaries.

---

Disclaimer: