Kalman-MTF Expansion Arb
Family: hybrid · Regime: ranging · Complexity: high · Asset classes: Crypto (ETH/BTC Pair), Forex (Major Crosses) · Timeframes: H1, H4
Thesis
The hypothesis is that asset pairs with a high degree of cointegration (like ETH/BTC) will periodically deviate from their fair-value spread due to temporary liquidity imbalances. These deviations are best captured when statistical extremes (Z-Score > 2) are met with a 'volatility expansion' (Inside Bar/Outside Bar hybrid) that signals the start of the mean-reversion move, provided that relative buyer strength (Bulls Power) confirms the direction. The edge exists because markets tend to overreact to news affecting one side of a pair before returning to the fundamental ratio.
Components
- AI Pair Trading System v1.0 (regime) — Establishes the mean-reversion regime by identifying when a pair spread deviates statistically from its Kalman-filtered equilibrium.
- EMA Dynamic Selector Test (direction) — Acts as a ultra-short-term trend filter to ensure the immediate momentum aligns with the expected mean-reversion direction.
- Inside Bars (MTF Framework) (entry) — Triggers entry on a volatility expansion (based on the provided formula High >= High[1] and Low <= Low[1]) which signals a breakout from the spread extreme.
- Stochastic Oscillator (exit) — Signals momentum exhaustion as the spread approaches the mean, providing a dynamic exit point.
- Price Line and Visible Range High-Low (risk) — Determines the hard stop-loss based on the visible chart structure (Visible Range Low for Longs/High for Shorts).
- Bulls Power (Elder Ray) (confirmation) — Confirms that the breakout is supported by physical buying pressure (High relative to its own EMA) rather than just noise.
Known failure conditions
- The spread between Asset A and Asset B enters a non-stationary 'walk' where the Kalman Mean fails to track price (breakdown of cointegration).
- Market volatility drops below the threshold required to trigger the 'Inside Bar' (Expansion) signal.
- Execution costs (spread/commissions) exceed the narrow profit margins of the mean-reversion move.
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).