KNN-DPO Cyclical Regime Follower

Family: hybrid · Regime: trending · Complexity: high · Asset classes: Equities, Crypto, Large-cap FX · Timeframes: H1, H4, D1

Thesis

Market regimes can be classified by comparing current price action to historical 'neighbors'. When this AI-derived regime aligns with medium-term momentum (MACD), cyclical pullbacks (DPO) represent high-probability trend-continuation entries. The edge relies on the assumption that price cycles mean-revert to a displaced moving average (DPO logic) before continuing in the direction of the non-linear regime (KNN).

Components

• AI Trend Navigator [K-Neighbor] (regime) — Provides the primary regime filter by identifying the 'nearest neighbor' price state to determine if the current market structure is historically associated with upward or downward movement.
• Connect MACD (Azullian) (direction) — Acts as a directional momentum filter to ensure the DPO cycle entries are aligned with broader MACD-derived momentum.
• Detrended Price Oscillator (DPO) (entry) — Identifies local cyclical lows (oversold) within an uptrend or local highs (overbought) within a downtrend for precision entry.
• ATR SL Finder (exit) — Used for dynamic trailing exit to protect profits and exit when volatility-adjusted price levels are breached.
• Average True Range (NNFX Version) (risk) — Calculates the initial stop loss distance and determines the position size based on current market volatility.
• Plotshape Test Indicator (confirmation) — A hardcoded price floor confirmation (Close > 2) that filters out low-value/penny-stock assets where volatility calculations become unreliable.

Known failure conditions

• The asset price falls below 2 permanently, rendering the confirmation indicator incapable of producing signals.
• Market volatility drops to levels where NNFX ATR provides stop distances smaller than the average spread.
• High-frequency regime switching where KNN neighbors provide contradictory predictions bar-to-bar.

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.