Recursive Logit Drift Trend-Follower
Family: trend_following · Regime: trending · Complexity: high · Asset classes: Crypto (ATOM/USDT preferred), Major FX Pairs · Timeframes: 15m, 1h
Thesis
Trend persistence is highest when (1) an asset exhibits non-random statistical drift, (2) the underlying currency shows strength across a broad basket, and (3) micro-structure factors like liquidity gaps (FVGs) align with the directional bias. By using a recursive smoothing trigger (RMAU) only when these conditions are met, we filter out noise-induced 'fakeouts' common in standard trend systems.
Components
- Currency Strength Template (regime) — Identifies if the base currency's strength is derived from broad market flow rather than an idiosyncratic spike in a single pair.
- Simple Moving Average (SMA) (direction) — Provides a secondary 'sanity check' trend filter to ensure the recursive entry isn't fighting a major timeframe's inertia.
- Recursive Moving Average Unlimited (entry) — Acts as the precise trigger; the recursive smoothing filters out high-frequency noise that typically triggers premature trend entries.
- SuperTrend (exit) — Provides a trailing stop-loss mechanism and a clear exit signal when the trend structure (ATR-based) breaks.
- ATOM Coefficient Indicator (logit) (risk) — Uses logistic regression to verify if the micro-structure (FVGs and Liquidity) supports a high-probability trade at the moment of entry.
- Swiss Army Knife Indicator (SAK) (confirmation) — Uses a BandPass filter to confirm that price momentum is within a specific cyclical frequency, reducing entries in stagnant markets.
- Asset Drift Model (ADM) (volatility_filter) — Ensures the asset is exhibiting statistically significant drift (non-random walk behavior) before committing capital.
Known failure conditions
- Asset Drift Model (ADM) t-stat remains between -1.96 and 1.96 for extended periods, indicating a random walk regime.
- ATOM Coefficient P-value stays near 0.50 regardless of price movement, indicating model obsolescence.
- SuperTrend flip frequency increases 3x above historical average, indicating a range-bound environment unsuitable for RMAU entries.
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.