KNN Supertrend Horizon [LuxAlgo]

The KNN Supertrend Horizon indicator is a machine learning tool that combines K-Nearest Neighbors (KNN) classification with Supertrend logic to identify high-probability trend directions and price rejection zones.

🔶 USAGE

The indicator provides a comprehensive view of market trends by filtering traditional Supertrend calculations through a machine learning engine. It is designed to help traders stay on the right side of the trend while identifying potential exhaustion points through visual "Rejection Orbs."

🔹 Trend Identification
The core trend logic is driven by the KNN engine, which analyzes RSI and volatility (ATR) features over a lookback window to determine the most likely trend direction. When the ML confidence aligns with the Supertrend, the indicator displays a colored horizon at the top or bottom of the chart.

🔹 3D Rejection Orbs
Specialized rejection signals appear when price interacts with the Supertrend level and forms a significant wick.

• The size of the orb is dynamically adjusted based on relative volume.
• A label attached to the orb displays the exact volume of the rejection candle.
• These orbs serve as potential entry or exit signals where price is "bouncing" off the trend baseline.

🔹 Confidence Visualization
The script features "Liquid Smooth" gradient candle coloring. The intensity of the candle colors shifts based on the ML engine's confidence level. Brighter, more vibrant colors indicate a high-conviction trend, while muted colors suggest the trend may be weakening or entering a sideways phase.

🔶 DETAILS

The KNN engine functions by searching for historical similarities in price behavior. By comparing the current RSI and ATR-based volatility to the past N bars (Search Window), it finds the K closest matches (Neighbors) to predict the current trend state. A confidence buffer is applied to this probability to filter out market noise and prevent rapid signal flipping.

🔶 SETTINGS

🔹 Machine Learning Settings

• K-Neighbors: The number of historical neighbors the algorithm looks for to determine the trend direction.
• Search Window: The lookback period (in bars) used to find similar historical patterns.

🔹 Supertrend Settings

• ATR Length: The period used for the Average True Range calculation.
• Factor: The multiplier applied to the ATR to set the distance of the Supertrend line.

🔹 Noise Filter Settings

• Smooth Price Input: Enables HMA smoothing on the price source used for ML features.
• ML Confidence Buffer (%): The percentage threshold above/below 50% required to trigger a trend change.

🔹 Rejection Signal Settings

• Show 3D Rejection Orbs: Toggles the visibility of the volume-based rejection bubbles.
• Min Wick-to-Body Multiplier: The required ratio of wick length to body size to qualify as a rejection.
• Min Bubble Gap: The minimum number of bars required between consecutive rejection signals.

🔹 Visual & Dashboard Settings

• Liquid Smoothness: Controls the EMA smoothing applied to the ML confidence for visual gradients.
• Vibrancy: Increases the color intensity of the gradient candles and horizons.
• Show Dashboard: Toggles the real-time statistics table containing trend direction, ML confidence, and relative volatility.