KNN-VPCI Statistical Trend Navigator

Family: hybrid · Regime: trending · Complexity: high · Asset classes: Equities, Forex, Crypto · Timeframes: H1, H4, D1

Thesis

Market trends are most persistent when institutional volume (VPCI) confirms the momentum of price patterns that resemble historical 'nearest neighbor' successes (KNN), provided the current price dispersion (MdAE) remains low enough to indicate a stable, non-chaotic transition.

Components

• AI Trend Navigator [K-Neighbor] (regime) — Defines the primary trend regime using non-linear nearest-neighbor smoothing to filter out localized noise.
• Momentum Oscillator (direction) — Ensures the price rate of change is positive (for longs) or negative (for shorts) relative to a 14-period benchmark.
• Frankenstein Ultimate Pro - ATM Logic (entry) — Provides the specific trigger via TEMA/LSMA crossover, filtered by an ATR-based volatility expansion requirement.
• Volume Price Confirmation Indicator (VPCI) (exit) — Signals the exit when volume no longer confirms price movement (divergence), suggesting trend exhaustion.
• Median Absolute Error (risk) — Calculates a robust volatility floor for stop-loss placement and position sizing, less sensitive to outlier spikes than standard deviation.
• Unsupported ta.macd Fast Length (Test Case) (confirmation) — Acts as a secondary momentum confirmation filter; requires the MACD line to align with the KNN regime.

Known failure conditions

• KNN prediction enters a 'flat' state where local neighbors yield no directional bias for >20 bars.
• VPCI remains consistently negative during a price rally, indicating a complete decoupling of volume from price.
• MdAE exceeds 5% of asset price, suggesting market chaos unsuitable for median-based error calculation.

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