RLS-KNN Adaptive Structure Trend

Family: trend_following · Regime: trending · Complexity: high · Asset classes: FX, Equities, Crypto · Timeframes: H1, H4

Thesis

Market trends are most tradable when short-term momentum (MACD) aligns with institutional volume (Volume SuperTrend) and structural 'cleanliness' (RLS Filter), provided price is protected by high-timeframe supply/demand zones. By requiring alignment across five distinct mathematical domains, we filter for high-conviction institutional moves.

Components

• RSIOMA (RSI of Moving Average) (regime) — Sets the macro-regime; only allows trades when the smoothed RSI is above/below the 50-level mid-point.
• Frankenstein Ultimate Pro - ATM Logic (direction) — Determines local trend direction using LSMA and ensures a minimum volatility threshold is met before entry.
• MACD Classic (3-Line) (entry) — Provides the final execution trigger when momentum shifts in the direction of the established regime.
• Equity Line Projection (exit) — Technically an interactive utility, it is used here to manage the trade exit by projecting the 1.5 RR target based on the S&D stop loss.
• Supply & Demand Zones (NNFX) (risk) — Identifies structural support/resistance to anchor the stop loss and determine position size based on proximity to institutional zones.
• Volume SuperTrend AI (Expo) (confirmation) — Uses KNN to classify if the current volume-weighted price action matches historical trend-continuation profiles.
• Recursive Least Squares Adaptive Filter (RLS) (volatility_filter) — Filters out white noise by ensuring price is successfully converging with the RLS filter, indicating a 'clean' trend.

Known failure conditions

• RSIOMA remains pinned above 80 or below 20 for extended periods while price moves inversely (momentum exhaustion).
• The P-matrix of the RLS filter becomes poorly conditioned, leading to vertical 'spikes' in the filter line.
• KNN classification produces alternating signals every 2-3 bars, indicating a lack of historical precedent for the current volatility.

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).