K-Nearest-Neighbor Adjustment of the Relative Strength Index

Summary

This indicator augments a conventional RSI with a k-nearest-neighbors calculation. It describes comparing current conditions with historical observations using features such as RSI, momentum, volatility, regression slope, and price momentum. Outcomes from the nearest observations are weighted by similarity, then blended with the standard RSI. The settings shown allow the user to choose the RSI period, smoothing method, neighbor count, historical lookback, feature count, and the weight assigned to the machine-learning component.

The indicator also offers additional smoothing, configurable overbought and oversold thresholds, and display modes that color the line according to trend or impulse conditions. The supplied source is truncated, so the full feature construction, neighbor selection, outcome labels, and blending calculation cannot be checked from this document. It provides no validation results, comparison with ordinary RSI, or evidence that the added model improves predictions. The method therefore describes an indicator design rather than a demonstrated trading edge.

Key ideas

• The indicator combines standard RSI with a similarity-weighted estimate based on historical neighbors.
• Its described comparison features include RSI, momentum, volatility, regression slope, and price momentum.
• Settings control the RSI calculation, smoothing, neighbor count, lookback length, feature count, and machine-learning blend weight.
• Thresholds and visual modes can mark overbought, oversold, trend, or impulse conditions.
• The truncated source and lack of validation results prevent checking the model details or assessing predictive value.

Tags

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