LSTM-Regime Volatility Breakout System

Family: hybrid · Regime: trending · Complexity: high · Asset classes: FX, Equities, Crypto · Timeframes: H1, H4, D1

Thesis

Market trends are most tradable when volatility is expanding and supported by cumulative volume flow. By using an LSTM-optimized SuperTrend to filter for high-quality regimes and Bollinger Bands to identify specific volatility breakouts, we can capture the meat of a move while using volume-based momentum (TSI_CD) to exit before mean reversion occurs. The 'edge' lies in the AI's ability to adjust the SuperTrend factor to current market noise levels.

Components

• ML SuperTrend (Ultimate) - Auto-Optimized AI with LSTM (regime) — Acts as the primary regime filter, using K-Means clustering and LSTM to distinguish between 'High Quality Trend' and 'Noise' based on non-linear price features.
• Unsupported matrix remove_col (direction) — While containing a semantic error in its diagnostic code, the underlying plot(close) provides the raw price-action directional bias required to prevent counter-trend entries.
• Bollinger Bands (Standard) (entry) — Triggers entries when price exceeds standard volatility thresholds, identifying 'volatility breakouts' within the ML-defined regime.
• TSI Convergence Divergence (TSI_CD) (exit) — Exits positions when the momentum histogram flips color, capturing the deceleration of the trend before a full reversal.
• Extrem SuperTrend (risk) — Provides the structural floor/ceiling for the trade, defining the hard stop loss and the dynamic trailing mechanism.
• Price and Volume Trend (PVT) (confirmation) — Confirms that the price breakout is supported by cumulative volume flow, filtering out low-liquidity spikes.

Known failure conditions

• ML model weights fail to converge or result in 'flat' regime classification (constant noise signal).
• Price oscillates strictly between the Bollinger Bands and the Extrem SuperTrend stop, inducing a 'stop-and-reverse' death spiral.
• PVT divergence persists for over 50 bars while price trends, indicating the volume component is decoupling from price discovery.

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.