Linear-Neural Momentum Convergence
Family: hybrid · Regime: trending · Complexity: high · Asset classes: Forex, Equities, Crypto · Timeframes: 15m, 1h, 4h
Thesis
The strategy hypothesizes that momentum signals are most reliable when the price path adheres to a high-fidelity linear regression model (low Standard Error). By using an ultra-fast MACD (1.5 period) for direction and a K-Nearest Neighbor algorithm for non-linear smoothing, we can enter 'clean' trends on stochastic pullbacks and exit at volatility extremes (Bollinger Bands) before the linear model breaks down.
Components
- Standard Error of Regression (STDERR) (regime) — Used as a regime filter to ensure price action is currently respecting a linear trend model with low dispersion before allowing entries.
- Unsupported ta.macd Fast Length (Test Case) (direction) — Provides ultra-sensitive momentum direction. The 1.5 fast length creates a near-instantaneous reaction to price change relative to the slow period.
- OTLIB Stochastic Oscillator (entry) — Identifies cyclical pullbacks within the established trend. LWMA smoothing reduces noise compared to standard SMA-based stochastics.
- Bollinger Bands (Standard) (exit) — Used for mean-reversion exits; closing positions when price reaches the statistical extremity of the current volatility envelope.
- ATR SL Finder (risk) — Provides a dynamic, volatility-adjusted stop loss based on the recent range to protect against sudden spikes.
- AI Trend Navigator [K-Neighbor] (confirmation) — Acts as a non-linear confirmation filter to ensure the ultra-fast MACD signal aligns with local price-pattern similarities.
Known failure conditions
- STDERR remains consistently high for extended periods, suggesting the market has moved into a non-linear or chaotic state where regression-based filtering is useless.
- The ultra-fast MACD (1.5 period) generates excessive whipsaws that the KNN and STDERR filters fail to mitigate.
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.