Evaluating Trading Systems by Testing Their Components Separately

Summary

The article proposes evaluating a complex trading system by measuring its component blocks separately with shared criteria, rather than optimizing every parameter at once. It uses net profit, maximum drawdown, and the share of profitable trades to compare two entry modules: Parabolic SAR and Accelerator Oscillator signals. The modules are tested individually and together, with a test mode selecting which configuration runs.

In the reported example, the combined system had higher net profit and a lower maximum drawdown than either single module. The author then changes the timeframe used by the Parabolic SAR block and compares the degraded component and whole-system results with the original configuration. The example suggests that isolating a weak block can guide targeted adjustment and ongoing monitoring. These are specific backtest results, however; the document does not establish robustness across markets or periods, and its conclusions depend on the chosen settings and test conditions.

Key ideas

• Shared performance criteria allow system blocks to be compared individually and as a combined system.
• The example measures net profit, maximum drawdown, and the proportion of profitable trades.
• A test mode runs Parabolic SAR, Accelerator Oscillator, or both entry modules.
• Changing the Parabolic SAR timeframe worsens its reported standalone metrics and the combined system metrics.
• Component-level analysis can focus optimization on underperforming modules, but the example does not establish broad robustness.

Tags

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