Adaptive SuperTrend Using K-Means Volatility Regimes

Summary

This indicator modifies SuperTrend by grouping recent ATR observations into low, medium, and high volatility regimes with a three-cluster K-Means procedure. It initializes cluster centers from percentiles of the ATR range over a training window, repeatedly assigns observations to the nearest center and recalculates each center, then uses the center nearest to current ATR to set the SuperTrend bands. A multiplier controls band distance, while the training length and iteration limits control adaptation and computation.

The chart display includes trend coloring, direction-change arrows, and a dashboard showing the assigned volatility regime and raw ATR. The document argues that regime-based bands may reduce sensitivity to brief volatility spikes while allowing tighter or wider stops in different conditions, but it provides no backtest or comparative results to support those claims. Clustering depends on the selected window and initialization, and the indicator remains a lagging, parameter-sensitive trend tool rather than a validated source of trading returns.

Key ideas

• The method clusters recent ATR values into three volatility regimes using K-Means.
• The nearest cluster center to current ATR is used to scale the SuperTrend bands.
• A training window, band multiplier, initialization percentiles, and convergence limits are configurable.
• Trend changes are marked with arrows, and the chart reports the current volatility classification.
• The document provides no empirical comparison showing that the adaptive version improves trading results.

Tags

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