Wavelet-ASI Logit Cyclical Momentum

Family: hybrid · Regime: trending · Complexity: high · Asset classes: Crypto (specifically ATOMUSDT) · Timeframes: 15m

Thesis

The strategy hypothesizes that ATOMUSDT 15m price action contains high-frequency noise that masks underlying cyclical trends. By using the Stationary Haar Discrete Wavelet Transform to denoised price and the Accumulation Swing Index to validate the 'true' OHLC swing structure, we can identify high-probability trend windows. The ATOM Logit model acts as a structural filter, entering only when local liquidity and FVGs favor the momentum, thereby overcoming the lag inherent in standard indicators.

Components

• Accumulation Swing Index (ASI) (regime) — Filters for the 'true' market direction by considering OHLC relationships, ensuring the trend has structural backing beyond simple close prices.
• Discrete Wavelet Transform (DWT) (direction) — Decomposes price into cyclical components; level 0 provides the denoised approximation of the current trend direction.
• ATOM Coefficient Indicator (logit) (entry) — Provides a non-linear entry trigger based on local market structure features (FVGs, liquidity) specific to the 15m ATOMUSDT pair.
• Parabolic SAR (Standard) (exit) — Provides a dynamic, trailing exit mechanism that tightens as the trend matures and exits on a definitive reversal.
• Gain Loss Info (Visual Overlays) (risk) — Uses historical candle percentage changes to calibrate position sizing based on realized volatility.
• Momentum Oscillator (confirmation) — Confirms that the velocity of price movement is aligned with the wavelet trend and logit prediction.

Known failure conditions

• The ATOM Logit model probability (P) stays centered between 0.45 and 0.55 for extended periods during high volatility.
• ASI 'Limit Move' T is set too high, causing the SI to remain flat in volatile crypto markets.
• Haar wavelet 'boxiness' creates significant lag, causing entries after the move is 50% complete.

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).