Combining Linear Regression Channels, Pivot Support, and an EMA Trend Filter

Summary

This long-only strategy seeks pullbacks within an upward bias. It builds a linear regression center line over a rolling price window and places upper and lower bands at multiples of recent standard deviation. The bias is defined by the upper band being above a 20-period exponential moving average. A potential entry occurs when price reaches the lower band or comes within a small buffer of the latest confirmed pivot low, with a bullish candle close.

The script marks pivot highs and lows as resistance and support, and exits if price reaches the upper band or the lower band is below the EMA. The concept is a blend of trend filtering and buying a pullback toward a statistical or horizontal support area. The document explains the rules but reports no test results, transaction-cost sensitivity, or comparison with alternatives. Pivot confirmation requires later bars, and the support buffer, channel settings, and exit conditions can materially affect signals and backtest behavior.

Key ideas

• The strategy defines an upward bias when the upper regression band sits above the 20-period EMA.
• The regression channel uses a rolling linear regression and standard deviation to frame price variation.
• Long entries require a pullback to the lower band or near the latest pivot support, plus a bullish candle.
• Positions close at the upper band or when the lower band is below the EMA.
• The document provides no performance evidence, and pivot confirmation introduces signal delay.

Tags

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