Adaptive Trend Filtering with Smoothed Price Ranges

Summary

This trend-following method builds a dynamic price filter from smoothed ranges. It calculates absolute price changes over fast and slow periods, smooths each with exponential moving averages, adjusts them with multipliers, and averages the resulting ranges. A range-filtering step limits how far the filtered price can move from its prior value. The system tracks consecutive rises or falls in that filtered price and derives adaptive upper and lower boundaries. A close crossing the trend filter generates a long or short entry signal.

The document presents the boundaries as volatility-responsive levels intended to reduce noise, but supplies no quantified tests or trading results. Its listed caveats include smoothing lag, repeated false crosses in sideways conditions, sensitivity to period and multiplier choices, and the absence of an explicit stop loss in the described implementation. It proposes additional confirmation, dynamic parameter adjustment, and risk controls as possible extensions. These are suggestions rather than demonstrated improvements, and the strategy’s performance across instruments or market regimes remains unestablished.

Key ideas

• Fast and slow smoothed ranges are combined to scale a dynamic price filter.
• The filter constrains changes in its value and supports adaptive upper and lower boundaries.
• Crosses of the closing price and trend filter generate directional entry signals.
• Smoothing may delay signals, while sideways markets can produce repeated false crosses.
• The document reports no performance results and notes that the described implementation lacks an explicit stop loss.

Tags

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