Empirical Mode Decomposition for Trend and Cycle Regimes

Summary

The document introduces Empirical Mode Decomposition (EMD) as an adaptive time-domain approach for breaking a complex signal into intrinsic mode functions. It describes EMD as suitable for nonlinear and nonstationary data, and notes that high-frequency components in the first mode can be removed for smoothing. These general properties are background; the trading indicator shown is a simplified filter rather than a full decomposition procedure.

The indicator applies a band-pass calculation to median price, smooths its output, identifies local peaks and valleys, and averages those turning points. A configurable fraction of the peak and valley averages sets upper and lower thresholds: the smoothed trend above or below them signals an uptrend or downtrend, while values between them indicate a cycle regime. Threshold spacing is subjective, and the author favors wider thresholds to focus on cycle trading and avoid it during clear trends. The page provides an indicator formula but no market data, backtest, or evidence of profitability, so its effectiveness and parameter robustness remain unestablished.

Key ideas

• EMD adapts signal decomposition to local time scales and can be applied to nonlinear, nonstationary processes.
• The indicator smooths a band-pass price signal and tracks local peaks and valleys.
• Fractions of averaged peaks and valleys define boundaries between trend and cycle regimes.
• The threshold fraction is subjective and should reflect the intended trading style.
• The page presents no empirical performance evidence for the indicator.

Tags

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