Kalman-Ehlers Adaptive Arbitrage (KEAA)

Family: hybrid · Regime: trending · Complexity: high · Asset classes: Equities, Crypto-Pairs (e.g., ETH/BTC), Forex Crosses · Timeframes: 1H, 4H, Daily

Thesis

Asset pair spreads exhibit mean-reverting behavior (detected by Kalman-Z-Scores) that is most tradable when the underlying price trend is confirmed by low-lag smoothing (Ehlers USF). By using GARCH(1,1) to estimate conditional volatility, we can adjust risk dynamically, assuming that volatility clusters and that mean reversion is less reliable during periods of regime-breaking variance. This hypothesis posits that the edge lies in the intersection of statistical arbitrage and adaptive risk management.

Components

• Ehlers Ultimate Smoother Filter (USF) (regime) — Defines the primary trend regime; its low-lag property allows for faster response to trend shifts compared to SMAs.
• Profile Dynamic Series Max Bars Back Test (direction) — Acts as a 3-bar 'memory' anchor to ensure the current price has moved significantly relative to its immediate past (momentum confirmation).
• AI Pair Trading System v1.0 (entry) — Provides the core mean-reversion trigger based on spread divergence between two assets, refined by Kalman filtering.
• Bears Power (exit) — Detects the exhaustion of selling pressure (long exit) or resurgence of it (short exit).
• Conditional Volatility (CV) (risk) — Predicts volatility clustering to dynamically adjust stop-loss width and position size based on current market risk.
• Bollinger Bands (Standard) (volatility_filter) — Acts as a 'tail-risk' filter; entries are inhibited if the price is already 'riding the bands' to avoid entering at local exhaustion points.

Known failure conditions

• Failure of the underlying asset correlation (cointegration breakdown) in the AI Pair Trading system.
• A 'volatility explosion' where GARCH(1,1) estimates remain at the upper bound, causing position sizes to approach zero.
• Persistent trend-less 'sawtooth' price action where USF toggles direction every few bars.

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.