Building a Modular Multi-Signal Expert Advisor in MQL5
Summary
The article presents a modular design for an Expert Advisor that combines custom and built-in trading signals through the MQL5 Standard Library and Wizard. Signal modules can be developed independently, assigned weights, and given different roles: primary signals propose trades, while filters qualify or block them. The proposed decision process totals signal confidence and compares it with a threshold before position sizing and execution.
As an implementation example, the author introduces a Fibonacci retracement signal module intended for Wizard integration and pairs it with Moving Average, RSI, and Accelerator Oscillator signals. The article describes the module’s configurable parameters and identifies source files for a standalone test EA and a multi-signal EA. However, the material does not establish that the combined system is profitable or demonstrate a validated consensus rule. It explicitly notes that Wizard-generated code does not make signals collaborate intelligently on its own; developers must refine that logic manually and test parameter choices.
Key ideas
- Separate signal logic into modules that can be developed and adjusted independently.
- Use weights and a confidence threshold to combine signal contributions into trade decisions.
- Assign primary signals and filters distinct roles in the system.
- A Fibonacci retracement module can be integrated with built-in MQL5 signals.
- Wizard generation supplies structure, but developers must implement and validate how signals interact.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.