OPEN-SOURCE SCRIPT
AetherEdge AI Divergence Ghost
🖊️ Overview
AetherEdge AI Divergence Ghost is a next-generation divergence detector that neurally fuses five oscillators and self-evolves their weights based on real-time correlation with future price action. By eliminating single-oscillator noise and filtering through a multi-layer confidence score, it plots only elite-grade signals as ultra-minimal ghost arrows — a silent sniper indicator built for clarity and conviction.
🔶 Key Features
Multi-Oscillator Neural Fusion: RSI, MACD, Stochastic, MFI, and CCI normalized via Z-score
Adaptive Weight Learning: Softmax weights derived from each oscillator's correlation with forward returns
4-Type Divergence Detection: Regular Bull/Bear & Hidden Bull/Bear
Confidence Scoring: Combines divergence magnitude with multi-oscillator agreement
Pivot Distance Filter: minBars / maxBars eliminate low-quality setups
Cooldown Mechanism: Prevents arrow clustering
Ghost Arrow Visuals: Tiny, transparency-adjustable triangles keep charts pristine
🧠 Technical Architecture
The engine combines a self-evolving neural oscillator with multi-layer validation.
Normalization Layer: All five oscillators are converted to a 50-period Z-score (x-μ)/σ, ensuring scale-uniform fusion.
Adaptive Weight Learning: A 100-bar correlation between each oscillator (lagged 5 bars) and forward return is computed. Only positive correlations contribute, normalized softmax-style — so oscillators that actually work in the current regime gain influence automatically.
Neural Composite Oscillator: neuralOsc = Σ(w_i × n_i) produces a regime-optimal signal.
Dual-Layer Pivot Detection: Pivots are tracked on both price and the neural oscillator, evaluating all four divergence types logically.
Confidence Scoring: Combines price/oscillator delta magnitude with an agreement bonus (how many of 5 raw oscillators confirm the divergence direction) into a final composite score.
Cooldown & Threshold Filter: Signals fire only when score ≥ minScore AND cooldown bars have elapsed — strict elite-only output.
⚙️ Recommended Settings & Tuning Guide
Crypto Defaults:
BTC/ETH (1H–4H): Lookback=50, Pivot Strength=5, Min Score=0.6
SOL/XRP (15M–1H): Pivot Strength=3, Min Score=0.65 (noise-resistant)
Swing (4H–1D): Lookback=80, maxBars=100, Min Score=0.55
Tuning Tips:
Too few signals: Lower Min Score to 0.5, or disable Adaptive Weight Learning
Too many signals: Raise Min Score to 0.7+, Pivot Strength to 7+
Trending regime: Show Hidden only (disable Regular)
Ranging regime: Show Regular only (disable Hidden)
💡 How to Use in Practice
Regular Bull/Bear: Counter-trend reversal entries
Hidden Bull/Bear: Trend-continuation pullback/rally entries
Multi-Timeframe: Use HTF Regular divergences for bias, LTF Hidden divergences for precision entries
Combinations: Pair with AetherEdge Adaptive Super Neural — Ghost signals coinciding with trend flips offer extreme conviction
S/R Confluence: Ghost arrows at major S/R zones dramatically boost reversal probability
⚠️ Important Notes
Pivots confirm pivotLen bars after formation — signals plot historically (confirmation delay, not repaint)
The first ~100 bars serve as a learning warm-up; Adaptive Weight accuracy improves thereafter
After abrupt volatility shifts, normalization needs several bars to restabilize
Divergence offers probabilistic edge — always combine with confluence and risk management
🚨 Disclaimer
This indicator is provided for educational and research purposes only and does not constitute financial advice. All trading decisions are made at your own risk and should be accompanied by proper risk management. Past performance is not indicative of future results.
AetherEdge AI Divergence Ghost is a next-generation divergence detector that neurally fuses five oscillators and self-evolves their weights based on real-time correlation with future price action. By eliminating single-oscillator noise and filtering through a multi-layer confidence score, it plots only elite-grade signals as ultra-minimal ghost arrows — a silent sniper indicator built for clarity and conviction.
🔶 Key Features
Multi-Oscillator Neural Fusion: RSI, MACD, Stochastic, MFI, and CCI normalized via Z-score
Adaptive Weight Learning: Softmax weights derived from each oscillator's correlation with forward returns
4-Type Divergence Detection: Regular Bull/Bear & Hidden Bull/Bear
Confidence Scoring: Combines divergence magnitude with multi-oscillator agreement
Pivot Distance Filter: minBars / maxBars eliminate low-quality setups
Cooldown Mechanism: Prevents arrow clustering
Ghost Arrow Visuals: Tiny, transparency-adjustable triangles keep charts pristine
🧠 Technical Architecture
The engine combines a self-evolving neural oscillator with multi-layer validation.
Normalization Layer: All five oscillators are converted to a 50-period Z-score (x-μ)/σ, ensuring scale-uniform fusion.
Adaptive Weight Learning: A 100-bar correlation between each oscillator (lagged 5 bars) and forward return is computed. Only positive correlations contribute, normalized softmax-style — so oscillators that actually work in the current regime gain influence automatically.
Neural Composite Oscillator: neuralOsc = Σ(w_i × n_i) produces a regime-optimal signal.
Dual-Layer Pivot Detection: Pivots are tracked on both price and the neural oscillator, evaluating all four divergence types logically.
Confidence Scoring: Combines price/oscillator delta magnitude with an agreement bonus (how many of 5 raw oscillators confirm the divergence direction) into a final composite score.
Cooldown & Threshold Filter: Signals fire only when score ≥ minScore AND cooldown bars have elapsed — strict elite-only output.
⚙️ Recommended Settings & Tuning Guide
Crypto Defaults:
BTC/ETH (1H–4H): Lookback=50, Pivot Strength=5, Min Score=0.6
SOL/XRP (15M–1H): Pivot Strength=3, Min Score=0.65 (noise-resistant)
Swing (4H–1D): Lookback=80, maxBars=100, Min Score=0.55
Tuning Tips:
Too few signals: Lower Min Score to 0.5, or disable Adaptive Weight Learning
Too many signals: Raise Min Score to 0.7+, Pivot Strength to 7+
Trending regime: Show Hidden only (disable Regular)
Ranging regime: Show Regular only (disable Hidden)
💡 How to Use in Practice
Regular Bull/Bear: Counter-trend reversal entries
Hidden Bull/Bear: Trend-continuation pullback/rally entries
Multi-Timeframe: Use HTF Regular divergences for bias, LTF Hidden divergences for precision entries
Combinations: Pair with AetherEdge Adaptive Super Neural — Ghost signals coinciding with trend flips offer extreme conviction
S/R Confluence: Ghost arrows at major S/R zones dramatically boost reversal probability
⚠️ Important Notes
Pivots confirm pivotLen bars after formation — signals plot historically (confirmation delay, not repaint)
The first ~100 bars serve as a learning warm-up; Adaptive Weight accuracy improves thereafter
After abrupt volatility shifts, normalization needs several bars to restabilize
Divergence offers probabilistic edge — always combine with confluence and risk management
🚨 Disclaimer
This indicator is provided for educational and research purposes only and does not constitute financial advice. All trading decisions are made at your own risk and should be accompanied by proper risk management. Past performance is not indicative of future results.
开源脚本
秉承TradingView的精神，该脚本的作者将其开源，以便交易者可以查看和验证其功能。向作者致敬！您可以免费使用该脚本，但请记住，重新发布代码须遵守我们的网站规则。
免责声明
这些信息和出版物并非旨在提供，也不构成TradingView提供或认可的任何形式的财务、投资、交易或其他类型的建议或推荐。请阅读使用条款了解更多信息。
开源脚本
秉承TradingView的精神，该脚本的作者将其开源，以便交易者可以查看和验证其功能。向作者致敬！您可以免费使用该脚本，但请记住，重新发布代码须遵守我们的网站规则。
免责声明
这些信息和出版物并非旨在提供，也不构成TradingView提供或认可的任何形式的财务、投资、交易或其他类型的建议或推荐。请阅读使用条款了解更多信息。