Combining Multi-Timeframe SuperTrend, KNN Trend Classification, and ADX

Summary

This strategy combines two SuperTrend signals calculated on separate timeframes with a K-nearest neighbors classifier and directional filters. The KNN component uses nearby historical SuperTrend observations to classify trend direction through weighted voting. A trade is considered when both timeframe signals agree and the ADX and directional movement index indicate a sufficiently strong trend. Users can choose long trades, short trades, or both, and may use a trailing stop for risk control.

The description gives example default parameter settings and refers to Bitcoin and Ether performance views, but it supplies no numerical results, test methodology, costs, or out-of-sample evidence. It therefore explains a signal design rather than establishing its effectiveness. Performance may depend on timeframe, market, parameter choices, and implementation details; the document advises adjusting inputs to asset conditions but does not demonstrate that this prevents overfitting or false signals.

Key ideas

• Two SuperTrend calculations on different timeframes provide separate trend inputs.
• KNN classifies each timeframe’s trend using weighted votes from nearby historical values.
• ADX and DMI filters are used to require trend strength and directional confirmation.
• The strategy supports long-only, short-only, or two-sided trading and mentions a trailing stop.
• The document describes settings and logic but provides no quantified validation evidence.

Tags

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