An advanced algorithmic trading system that combines NEAT (NeuroEvolution of Augmenting Topologies) with technical and fundamental analysis to create evolutionary and adaptive trading strategies.

• 🧠 NEAT Algorithm: Neural network evolution with dynamic topology
• 📊 Technical Analysis: 15+ technical indicators (MACD, ADX, Stochastic, CCI, MFI, ATR, etc.)
• 📰 Fundamental Analysis: Economic data processing and news sentiment analysis
• 🔄 Advanced Backtesting: Complete system with comprehensive performance metrics
• 📈 Real-time Visualization: Graphical interface for evolution tracking
• 🎯 Multi-Objective Optimization: Simultaneous optimization of return, risk, and consistency
• 🚀 Modular Architecture: Clean separation of concerns with src/ organization
• 🧪 Comprehensive Testing: Automated test suite with coverage reporting
• ⚡ Performance Optimized: Parallel processing and GPU acceleration support

src/
├── config/           # Centralized configuration
│   └── index.js     # Main config with NEAT, trading, and system settings
├── core/            # Core components
│   ├── TradingEnvironment.js  # Main trading simulation environment
│   ├── neat.js      # NEAT algorithm implementation
│   └── population.js # Population management
├── genetics/        # NEAT genetic algorithm
│   ├── Genome.js    # Individual neural network genome
│   ├── Population.js # Population evolution and management
│   ├── Species.js   # Species classification
│   ├── ConnectionGene.js # Connection gene structure
│   ├── Node.js      # Neural network nodes
│   └── Player.js    # Individual trading agents
├── indicators/      # Technical analysis
│   └── TechnicalIndicators.js # 15+ technical indicators
├── analysis/        # Fundamental analysis
│   └── FundamentalAnalysis.js # Economic data processing
├── data/           # Data providers
│   └── DataProvider.js # Market data fetching and management
├── visualization/   # Real-time visualization
│   ├── TradingVisualizer.js # Main visualization engine
│   └── visualizer.js # Additional visualization utilities
└── tests/          # Comprehensive test suite
    ├── run-all-tests.js # Test runner
    ├── TradingEnvironment.test.js # Environment tests
    └── TechnicalIndicators.test.js # Indicator tests

Root Level:
├── BacktestEngine.js # Standalone backtesting engine
├── scripts/         # Utility scripts
│   └── run-backtest.js # Backtesting automation
├── libraries/       # Frontend libraries (p5.js, Box2D.js)
└── tests/          # Additional test files

git clone https://github.com/your-username/neat-trading-system.git
cd neat-trading-system

npm install

cp .env.example .env
# Edit .env with your API keys

# Console mode (automatic evolution)
npm start

# Development mode (with hot reload)
npm run dev

# Web interface
npm run web

# Development web server
npm run web:dev

• Node.js: >= 16.0.0
• Memory: Minimum 4GB RAM (8GB+ recommended for large populations)
• Storage: 1GB+ for historical data and logs

• MACD (Moving Average Convergence Divergence) - Signal, histogram, and crossovers
• ADX (Average Directional Index) - Trend strength with +DI/-DI
• EMA (Exponential Moving Average) - Multiple periods supported

• Stochastic Oscillator - %K and %D with customizable smoothing
• CCI (Commodity Channel Index) - Overbought/oversold detection
• RSI (Relative Strength Index) - Momentum oscillator

• OBV (On-Balance Volume) - Volume-price trend analysis
• MFI (Money Flow Index) - Volume-weighted RSI

• ATR (Average True Range) - Volatility measurement
• Keltner Channels - Volatility-based bands
• Bollinger Bands - Statistical price channels

• Double Top/Bottom - Reversal pattern detection
• Head and Shoulders - Classic reversal patterns
• Triangle Patterns - Continuation/reversal formations
• Market Regime Detection - Trend/range/volatile state identification

• Price Pattern Analysis - Advanced pattern recognition
• Market Regime Classification - Multi-factor regime detection
• Linear Regression - Trend analysis and slope calculation

// src/config/index.js
neat: {
    populationSize: 150,
    inputs: 20,  // Technical indicators + fundamental data
    outputs: 3,  // Buy, Sell, Hold
    mutationRates: {
        weight: 0.1,
        connection: 0.05,
        node: 0.02,
        enable: 0.1,
        disable: 0.1
    },
    speciation: {
        compatibilityThreshold: 3.0,
        excessCoeff: 1.0,
        disjointCoeff: 1.0,
        weightDiffCoeff: 0.4
    }
}

trading: {
    symbols: ['EURUSD', 'GBPUSD', 'USDJPY', 'AUDUSD'],
    timeframes: ['5m', '15m', '1h', '4h'],
    riskManagement: {
        maxRiskPerTrade: 0.02,
        maxDrawdown: 0.15,
        positionSizing: 'kelly'
    },
    backtesting: {
        startDate: '2020-01-01',
        endDate: '2024-01-01',
        initialBalance: 10000,
        commission: 0.0001
    }
}

data: {
    providers: {
        forex: 'alpha_vantage',
        news: 'newsapi',
        economic: 'fred'
    },
    apiKeys: {
        alphaVantage: process.env.ALPHA_VANTAGE_KEY,
        newsApi: process.env.NEWS_API_KEY,
        fred: process.env.FRED_API_KEY
    },
    updateIntervals: {
        prices: 60000,      // 1 minute
        news: 300000,       // 5 minutes
        economic: 3600000   // 1 hour
    }
}

# Run all tests
npm test

# Specific test suites
npm run test:trading
npm run test:indicators

# Code quality
npm run lint
npm run format

# Build for production
npm run build

The system includes comprehensive tests for:

• Trading Environment: Market simulation and state management
• Technical Indicators: All 15+ indicators with edge cases
• NEAT Algorithm: Genome evolution and population dynamics
• Backtesting Engine: Performance metrics and trade execution
• Data Providers: API integration and synthetic data generation

# Simple backtest
npm run backtest

# Specific backtest
node scripts/run-backtest.js EURUSD 5m 2023-01-01 2023-12-31

# Multiple backtests with export
npm run backtest -- --export

# Performance analysis
npm run analyze

# Parameter optimization
npm run optimize

�  Backtest Results:
   � Ionitial Balance: $10,000.00
   💰 Final Balance: $12,450.00
   📈 Total Return: 24.50%
   🎯 Win Rate: 67.3%
   📉 Max Drawdown: 8.2%
   📊 Sharpe Ratio: 1.847
   🔄 Total Trades: 156

• Multi-Symbol Testing: Test across multiple currency pairs simultaneously
• Multi-Timeframe Analysis: Compare performance across different timeframes
• Walk-Forward Optimization: Progressive parameter optimization
• Monte Carlo Simulation: Statistical robustness testing
• Performance Attribution: Detailed trade-by-trade analysis

O sistema otimiza simultaneamente:

1. Retorno Total (peso: 40%)
2. Taxa de Acerto (peso: 25%)
3. Sharpe Ratio (peso: 20%)
4. Drawdown Máximo (peso: 15%)

fitness = (totalReturn * 100) + 
          (winRate > 0.5 ? (winRate - 0.5) * 50 : 0) +
          (sharpeRatio * 10) -
          (maxDrawdown > 0.1 ? (maxDrawdown - 0.1) * 100 : 0)

O sistema inclui visualização em tempo real com:

• Gráfico de Preços: Candlesticks com indicadores
• Painel de Indicadores: MACD, ADX, Stochastic
• Informações da Conta: Balance, posições, trades
• Curva de Equity: Performance ao longo do tempo
• Distribuição P&L: Histograma de lucros/perdas

# Análise de performance
npm run analyze

# Otimização de parâmetros
npm run optimize

# Linting e formatação
npm run lint
npm run format

# Build para produção
npm run build

• Alpha Vantage: Dados de preços forex
• News API: Notícias financeiras
• FRED: Dados econômicos do Federal Reserve

O sistema gera dados sintéticos quando APIs não estão disponíveis, permitindo desenvolvimento e testes offline.

const NEATTradingSystem = require('./src/index.js');

const system = new NEATTradingSystem({
    neat: { populationSize: 100 },
    trading: { symbols: ['EURUSD'] }
});

await system.initialize();
const bestAgent = await system.runEvolution();

console.log('Best agent performance:', bestAgent.performance);

O sistema processa:

• Dados Econômicos: PIB, inflação, emprego, taxas de juros
• Calendário Econômico: Eventos futuros e impactos
• Análise de Sentimento: Processamento de notícias
• Diferenciais de Moedas: Comparação entre países

• Position Sizing: Kelly Criterion ou fixed fractional
• Stop Loss: Baseado em ATR ou percentual
• Drawdown Control: Parada automática em perdas excessivas
• Diversificação: Múltiplos pares e timeframes

• Deep Learning Integration: Modelos LSTM/Transformer
• Real-time Trading: Conexão com brokers
• Portfolio Management: Otimização multi-ativo
• Cloud Deployment: Execução em AWS/GCP
• Mobile App: Interface mobile
• Social Trading: Compartilhamento de estratégias

1. Fork o projeto
2. Crie uma branch para sua feature (git checkout -b feature/AmazingFeature)
3. Commit suas mudanças (git commit -m 'Add some AmazingFeature')
4. Push para a branch (git push origin feature/AmazingFeature)
5. Abra um Pull Request

Este projeto está licenciado sob a MIT License - veja o arquivo LICENSE para detalhes.

• Kenneth Stanley: Criador do algoritmo NEAT
• Comunidade QuantConnect: Inspiração em trading algorítmico
• TradingView: Referência em análise técnica

• 📧 Email: support@neat-trading.com
• 💬 Discord: NEAT Trading Community
• 📖 Wiki: Documentação Completa

⚠️ Aviso Legal: Este sistema é para fins educacionais e de pesquisa. Trading envolve riscos significativos. Sempre teste estratégias em ambiente simulado antes de usar capital real.