Combining Regression Error Bands with Bollinger Bands

Summary

This indicator combines two bands calculated over the same rolling window. The statistical component places upper and lower boundaries around a simple moving average using a multiple of price standard deviation, following the familiar Bollinger Band construction. The regression component fits an ordinary least squares line to recent closing prices and places boundaries around that line using a multiple of the standard error of the estimate. The supplied example sets the window to 21 bars and the multiplier to 2.

The author suggests that the regression-based bands may respond more quickly to price moves and volatility, while the standard deviation bands may help confirm medium- to longer-term movement. The document provides an indicator formula and a rationale, but no chart, trading rules, backtest, or performance evidence. It does not specify how to act on band crossings or account for market, instrument, or timeframe differences, so the claimed distinction should be treated as a hypothesis rather than a demonstrated edge.

Key ideas

• The indicator combines rolling standard deviation bands around a simple moving average with bands around a linear regression line.
• Regression bands use the standard error of the estimate, while statistical bands use price standard deviation.
• The example uses a 21-bar lookback and a multiplier of 2.
• The author proposes that regression bands react faster and standard deviation bands help confirm longer moves.
• No entry rules, backtest, or performance evidence are provided.

Tags

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