Adaptive Linear Regression Channels for Breakout and Range Trading

Summary

This document explains a rolling linear-regression channel built from recent closing prices. A least-squares line estimates the central path, and bands are placed above and below it using multiples of the residual standard deviation. The strategy uses price crossings of those bands for entries; in its default mode it closes positions when price crosses the center line. An alternate range mode changes the entry crossings. The published defaults specify a 100-bar window and one standard-deviation multiplier for each band.

The article claims some markets showed better backtests than moving-average approaches and refers to favorable live use, but supplies no results or supporting detail. It cautions that large price swings, poorly aligned channels, and the gap between historical tests and live trading can undermine performance. The source also uses the opening price for crossing signals, although the explanation emphasizes closing prices, and its center-band calculation may not match the described regression midpoint. Parameter tuning, added filters, stops, and position sizing are suggested, with overfitting remaining a concern.

Key ideas

• A rolling least-squares fit forms a central line, with bands scaled by residual standard deviation.
• Price crossings of the outer bands generate entries, while the default mode exits near the center.
• The source offers a range mode that changes which band crossings trigger entries.
• The document provides positive performance claims but no figures or supporting test details.
• The stated closing-price method differs from the source's use of opening prices for signal crossings.

Tags

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