Combining and Walk-Forward Optimizing Multiple Trading Strategies

Summary

This article develops a framework for comparing several trading strategies and potentially selecting among them over time. It introduces object-oriented MQL5 classes with a shared parent interface for updating indicators and generating buy and sell signals, allowing strategy-specific rules to be tested separately. The installment focuses on a moving-average continuation strategy: it compares moving averages of open and close prices, interpreting their relative position as directional signals. RSI momentum and Williams %R breakout approaches are introduced as additional strategies for later combination.

The article uses MetaTrader 5’s strategy tester and genetic optimizers to tune parameters, and explains walk-forward testing as a way to assess optimized settings beyond the training segment. It shows an equity curve with training and forward periods and states that both segments trend positively, alongside optimized settings. The material presents a testing workflow and one strategy example, rather than proof that dynamic strategy selection across all three strategies is profitable. Results depend on the selected instrument, timeframe, parameter search, and test period.

Key ideas

• A shared parent class lets strategy variants implement their own signal rules while reusing common methods.
• The moving-average example compares averages of open and close prices to generate directional signals.
• The article introduces RSI momentum and Williams %R breakout strategies for a broader multi-strategy system.
• Genetic optimization and walk-forward testing are used to search settings and evaluate them on a forward segment.
• The reported positive equity curves do not establish profitability across other markets or periods.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.