Kalman Filter Volatility Band Breakout Strategy

Summary

This document presents a trend-following system built around a two-state Kalman filter that estimates price and velocity. It measures the absolute difference between closing price and the filtered estimate, smooths that error over a lookback window, and places bands at a configurable multiple of the smoothed error. A close crossing above the upper band triggers a long entry; a close crossing below the lower band triggers a short entry, reversing the position as signals change.

The script includes example settings described as tuned for ETH on a 15-minute chart, and the accompanying text refers to a backtest spanning more than 200,000 candles. It supplies no reported returns, drawdowns, costs beyond the configured commission, or validation across other assets and periods. The parameter tuning claim therefore does not establish robustness, and the always-in-market reversal logic may be exposed to whipsaws and trading costs.

Key ideas

• A Kalman filter estimates price and velocity from closing prices.
• Smoothed absolute filter error sets the width of upper and lower volatility bands.
• Crossing above the upper band opens a long position, while crossing below the lower band opens a short position.
• The example is described as tuned for ETH on a 15-minute timeframe, but no performance statistics are provided.

Tags

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