KNN Volume Normality Trend Hunter

Family: hybrid · Regime: trending · Complexity: high · Asset classes: Equities, Indices · Timeframes: H1, D1

Thesis

Market trends are most exploitable when price returns deviate from a normal distribution (fat tails) and are confirmed by high-volume efficiency (Volume to Absolute Return Ratio). By using KNN to classify these volume-price relationships within a specific historical price corridor (Probable High/Low), we can identify high-conviction entries that represent non-random market movements.

Components

• Simple Moving Average (SMA) (regime) — Provides the primary trend filter; trades are only taken in the direction of the long-term baseline.
• ZigZag Color (Classic) (direction) — Identifies the current market structure leg to ensure alignment with immediate swing momentum.
• Volume SuperTrend AI (Expo) (entry) — Uses KNN classification to identify high-probability trend entries based on volume-weighted historical patterns.
• Volume Price Confirmation Indicator (VPCI) (exit) — Signals the end of a trend when volume no longer supports price movement, triggering an exit.
• Jarque-Bera Test (JB) (risk) — Determines position sizing by measuring the deviation from normality; higher JB values suggest non-random, trending behavior.
• Volume to Absolute Return Ratio (confirmation) — Confirms that price movement is backed by sufficient 'effort' (volume) relative to the return magnitude.
• Probable High Low (volatility_filter) — Acts as a hardcoded 'fair value' corridor. Trading only occurs when price is within these specific boundaries to avoid overextension.

Known failure conditions

• Price stays outside the hardcoded Probable High/Low levels for extended periods, rendering the system inactive.
• High frequency of ZigZag repainting leads to whipsaws in low-volatility environments.
• The KNN algorithm fails to generalize if market microstructure shifts (e.g., move to dark pools).

Explore the full interactive blueprint with parameter ranges and evidence on WOBR StrategyVerse, or generate this strategy as an MT4/MT5 Expert Advisor with QuantMogul AI Engine (free download).