Combining SuperTrend and KNN for Adaptive Trend Signals

Summary

This strategy combines a VWMA and ATR based SuperTrend with a weighted k-nearest-neighbors classifier. The SuperTrend bands use the moving average as a center and adjust their distance according to volatility. The classifier compares recent SuperTrend values with the current value, weights nearby observations by inverse distance, and uses their labels to estimate bullish or bearish conditions. The strategy enters when a trend begins or continues with a confirming KNN prediction, and uses dynamic trailing stops to exit. Traders can configure the trade direction and indicator parameters.

The listing mentions BTCUSD and ETHUSD performance on eight-hour charts, but provides no performance figures, test period, or comparison results. It therefore does not establish that the method is profitable or more accurate than a conventional SuperTrend. The description also gives limited detail about validation, position sizing beyond the listed defaults, or how the classifier behaves across market regimes. The stated strategy logic is best treated as a design outline that requires independent testing, including realistic trading costs.

Key ideas

• The SuperTrend bands are centered on a VWMA and scaled by ATR to reflect volatility.
• The KNN classifier weights the labels of nearby recent SuperTrend observations by inverse distance.
• Long and short signals combine new or continuing SuperTrend directions with the classifier's prediction.
• The strategy uses dynamic trailing stops and allows traders to choose long, short, or both directions.
• The listing names crypto chart examples but supplies no quantitative performance evidence or validation details.

Tags

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