Skip to content

Claude Quant - The Princeton Anomaly | AI-driven systematic futures trading | 2003 Princeton thesis | +54% live (32 days) | +240% CAGR (4yr backtest) | Time-zone arbitrage + VIX-adaptive risk

License

Notifications You must be signed in to change notification settings

ClaudeQuant/claude-quant

Folders and files

NameName
Last commit message
Last commit date

Latest commit

Β 

History

13 Commits
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 
Β 

Repository files navigation

Claude Quant Banner

Claude Quant

Live Performance Backtest License Python Pine Script


The Princeton Anomaly

In 2003, four Princeton graduate students discovered a systematic inefficiency in global futures markets. Twenty-one years later, one finally built itβ€”powered by Claude AI.

AI discovered it. AI coded it. AI trades it.


πŸ“Š Verified Performance

Live Trading (32 Days)

  • Total Return: +54.03%
  • Average Daily: +1.69%
  • Win Rate: 62.5%
  • Sharpe Ratio: 3.2+

Backtest (4 Years: 2021-2026)

  • CAGR: 240%
  • Total Return: +13,508% ($250K β†’ $34M)
  • Sharpe Ratio: 2.03
  • Max Drawdown: -35.32%

All data independently verifiable in /data folder.


πŸͺ™ $CQNT Token

πŸš€ Launching Q1 2026: First cryptocurrency token backed by audited systematic trading returns

Token Flywheel Mechanics

Trading Profits β†’ Buybacks β†’ Burns β†’ Price Growth

The $CQNT token creates value through a systematic flywheel:

  1. Token Launch β†’ Pump.fun (Solana) generates creator fees
  2. Investment β†’ Fees invested in Claude Quant strategy (240% CAGR)
  3. Profits β†’ Monthly trading returns compound
  4. Buybacks β†’ Profits used to purchase $CQNT tokens
  5. Burns β†’ Tokens permanently removed from supply
  6. Loop β†’ Scarcity + proven returns = price appreciation

Phased Buyback Schedule

  • Months 1-3: 100% of profits β†’ buybacks (ignition)
  • Months 4-6: 50% buybacks / 50% reinvest (momentum)
  • Months 7-9: 25% buybacks / 75% reinvest (transition)
  • Month 10+: 10% buybacks / 90% reinvest (perpetual)

Why this works: By Month 24, 10% of a $1.8M fund generates larger buybacks than 100% of initial $100K.

Complete Documentation

πŸ“„ Read the Full White Paper

Comprehensive explanation including:

  • 36-month financial projections with tables
  • Smart contract architecture details
  • Risk management framework
  • Transparency commitments
  • Monthly audit schedule

Key Differentiators

βœ… Real Revenue - Backed by proven 240% CAGR strategy
βœ… Complete Transparency - Daily GitHub updates + monthly CPA audits
βœ… Sustainable Model - Phased optimization for long-term growth
βœ… First of its Kind - No other token has this backing

Launch Details: Q1 2026 | Platform: Pump.fun (Solana) | Contract: TBA


πŸ€– The AI Implementation

Claude AI analyzed the original 2003 Princeton thesis and:

  1. Extracted the core time-zone arbitrage principle
  2. Coded the complete trading framework
  3. Enhanced it with modern VIX-adaptive risk management
  4. Deployed it for autonomous trading

The result: A 21-year-old academic insight, now executed by AI.


🧠 The System

Session Sequencing (Time-Zone Arbitrage)

  • Nikkei 225 (Asia) β†’ DAX (Europe) β†’ Nasdaq (US)
  • Same capital traded 3x per day
  • Zero overnight exposure

VIX-Adaptive Risk Management

  • Dynamic position sizing based on market volatility
  • Scales down in crisis, scales up when calm

Conditional Expansion

  • Profitable Asia β†’ Expanded Europe/US limits
  • Asymmetric risk-taking

Multi-Layer Protection

  • Portfolio hard stops
  • Per-market limits
  • Session controls
  • Automatic flattening

πŸ” Transparency & Verification

βœ… Publicly Available:

  • Complete CSV performance data
  • Python verification scripts
  • TradingView Pine Script strategy
  • Framework architecture code
  • Comprehensive documentation

⚠️ Proprietary:

  • Exact position sizing parameters
  • Signal generation algorithms
  • Conditional expansion values

Don't trust. Verify. Download and check yourself.


πŸš€ Quick Start

git clone https://github.com/ClaudeQuant/claude-quant.git
cd claude-quant
pip install pandas numpy matplotlib
python scripts/verify_performance.py

πŸ“ˆ Performance Notes

Why 240% CAGR:

Our backtest shows 240% Compound Annual Growth Rate (CAGR) over 4 years. This is the geometric mean - the actual rate at which capital compounded.

The Math:

  • Starting Capital: $250,000
  • Ending Value: $34,019,317 (after 4.01 years)
  • Total Return: +13,508%
  • CAGR: 240.12%

Formula: (Ending/Starting)^(1/Years) - 1

240% CAGR means doubling your money every 4.3 months on average.


πŸ“š Documentation

Trading Strategy

$CQNT Token


πŸŽ“ Academic Foundation

Built on a Princeton University thesis from 2003 exploring time-zone arbitrage. 21 years later, Claude AI has coded and deployed it with modern risk management.


πŸ“± Community


⚠️ Disclaimer

Trading futures involves substantial risk. 240% CAGR backtest with -35% max drawdown demonstrates significant volatility. Past performance does not guarantee future results.

$CQNT Token: Utility token, not a security. No guaranteed returns. See full disclaimers in white paper.

This is not investment advice.


The most transparent AI trading project in cryptocurrency.

21 years of academic research. Claude AI execution. Complete transparency.

Don't trust. Verify. πŸ”¬

About

Claude Quant - The Princeton Anomaly | AI-driven systematic futures trading | 2003 Princeton thesis | +54% live (32 days) | +240% CAGR (4yr backtest) | Time-zone arbitrage + VIX-adaptive risk

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages