Sigmoid-RAE Logistic Trend Follower
Family: trend_following · Regime: trending · Complexity: medium · Asset classes: Equities, Forex, Crypto · Timeframes: H1, H4, D1
Thesis
The strategy assumes that market trends do not move linearly but follow a logistic S-curve growth pattern. By using a Sigmoid function to isolate the 'meat' of the curve (the transition from 0 to 1) and filtering for high-conviction momentum (RAE and RSIOMA), we can enter trends as they accelerate and exit as soon as seller strength (Bears Power) begins to penetrate the trend's defensive moving average. This non-linear approach attempts to mitigate the late-entry problem of standard moving averages.
Components
- Logistic Function (SIGMOID) (regime) — Acts as a regime filter to ensure the current price action is in an 'expansion' phase of the logistic growth curve (S-curve), filtering out exhausted trends.
- Zero Lag Least Squares Moving Average (ZLSMA) (direction) — Provides the core trend bias with minimal lag, ensuring entries are aligned with the immediate directional flow.
- RSIOMA (RSI of Moving Average) (entry) — Identifies high-probability momentum entries by calculating relative strength on smoothed data, reducing false oscillations common in standard RSI.
- Bears Power (exit) — Detects when the 'bears' successfully push price below the trend EMA, serving as an early warning of trend exhaustion for exit.
- ATR Heiken Ashi (risk) — Uses smoothed HA candles to calculate a volatility-adjusted stop loss that is less sensitive to intraday noise than standard ATR.
- Relative Absolute Error (RAE) (volatility_filter) — Filters out low-volatility/mean-reverting environments where error-to-mean ratios are too low to support trend following.
Known failure conditions
- The RAE stays consistently below 0.4, indicating price is orbiting the mean without directional conviction.
- Sigmoid output saturates at 1.0 or 0.0 for extended periods while price continues to trend, rendering the regime filter useless.
- Price frequently oscillates around the ZLSMA in a tight range, causing consecutive stop-outs via the ATR HA buffer.
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.