Backtrader Framework for Brent Crude Oil Futures

The following is a simple backtesting framework for Brent Crude Oil Futures written with the backtrader library.

Description

This project queries historical data for Brent Crude Oil futures from Quandl and backtests several algorithmic trading strategies using the backtrader library. It currently ships with three strategies: SmaCrossoverStrategy, RSIMeanReversionSystem and ThreeSoldiersAndCrows, which were variously inspired by examples from the backtrader documentation series, backtest-rookies.com, as well as my own experiences in technical analysis.

The current implementation of this project is separated into two main functionalities: visualize_bt.py and optimize_bt.py.

Backtest Visualization

To perform a simple analysis using one of the pre-packaged strategies, simply edit in visualize_bt.py:

import extensions.strategies as st

cerebro.addstrategy(st.RSIMeanReversionSystem)

Then, in Terminal, run:

python visualize_bt.py

This returns a terminal output of P&L generated over a stipulated timeframe alongside the packaged backtrader plot.

Backtest Optimization

To optimize parameters within an existing strategy, first edit in optimize_bt.py:

cerebro.optstrategy(
        st.SmaCrossoverStrategy,
        sma1_period=range(10, 21),
        opt_mode=True
    )

Then run the following in Terminal:

python optimize_bt.py

This prints a series of simulations for the parameter(s) under analysis alongside their corresponding final portfolio values.

Future Plans (Ordered By Priority)

• Add to the set of strategies based on new ideas from fundamental research
• Include a plotting functionality for optimizations runs (i.e. Plotting Parameter Values against P&L)
• Analyze the potential of implementing the framework for live trading with OANDA or Interactive Brokers
• Tinker with 'extensions.sizers' to execute based on various risk portfolios
• Tinker with 'extensions.queries' to experiment with more unconvential commodity datasets