Spaces:
Configuration error
Configuration error
| # 🤖 Financial AI Predictor | |
| An advanced machine learning system for stock price prediction using ensemble models and comprehensive feature engineering. | |
|  | |
|  | |
|  | |
|  | |
| ## 🎯 Overview | |
| Financial AI Predictor combines state-of-the-art machine learning models with advanced financial feature engineering to predict stock prices. Built with robustness and educational value in mind, it demonstrates best practices for applying ML to financial markets. | |
| ### ✨ Key Features | |
| - 🧠 **Ensemble ML Models**: Random Forest + Gradient Boosting + Ridge Regression | |
| - 📊 **Advanced Feature Engineering**: 50+ technical indicators and market features | |
| - ⏰ **Real-time Data**: Live market data from Yahoo Finance | |
| - 🔬 **Robust Validation**: Time series cross-validation with walk-forward analysis | |
| - 🎯 **Uncertainty Quantification**: Confidence intervals and model consensus | |
| - 📈 **Interactive Interface**: Professional Gradio web interface | |
| - 🆓 **No API Keys**: 100% free public data sources | |
| ## 🚀 Quick Start | |
| ### Installation | |
| ```bash | |
| # Clone the repository | |
| git clone https://github.com/yourusername/financial-ai-predictor.git | |
| cd financial-ai-predictor | |
| # Install dependencies | |
| pip install -r requirements.txt | |
| # Run the application | |
| python financial_ai_predictor.py | |
| ``` | |
| ### Docker Setup (Optional) | |
| ```bash | |
| # Build Docker image | |
| docker build -t financial-ai-predictor . | |
| # Run container | |
| docker run -p 7860:7860 financial-ai-predictor | |
| ``` | |
| ### Basic Usage | |
| 1. **Launch the app**: `python financial_ai_predictor.py` | |
| 2. **Open browser**: Navigate to `http://localhost:7860` | |
| 3. **Enter stock symbol**: e.g., "AAPL", "GOOGL", "TSLA" | |
| 4. **Click "Run AI Analysis"**: Wait for model training and prediction | |
| 5. **Review results**: Analysis report, charts, and feature importance | |
| ## 📊 How It Works | |
| ### 1. Data Collection | |
| - Downloads 2 years of historical OHLCV data from Yahoo Finance | |
| - Real-time price and volume information | |
| - Automatic data quality checks and cleaning | |
| ### 2. Feature Engineering | |
| ```python | |
| # Technical Indicators | |
| - Price momentum (5, 20 day) | |
| - Moving averages (5, 10, 20, 50 day) | |
| - RSI, MACD, Bollinger Bands | |
| - ADX, Stochastic, Williams %R | |
| - Volume indicators (OBV, CMF, VWAP) | |
| # Market Regime Features | |
| - Volatility ratios and regimes | |
| - Trend strength indicators | |
| - Gap analysis and fill rates | |
| - Price-volume relationships | |
| ``` | |
| ### 3. Model Architecture | |
| ```python | |
| # Ensemble Models | |
| models = { | |
| 'random_forest': RandomForestRegressor(n_estimators=200), | |
| 'gradient_boost': GradientBoostingRegressor(n_estimators=150), | |
| 'ridge': Ridge(alpha=1.0) | |
| } | |
| # Time Series Validation | |
| validation = TimeSeriesSplit(n_splits=5) | |
| ``` | |
| ### 4. Prediction Pipeline | |
| 1. **Feature Extraction**: Calculate all technical indicators | |
| 2. **Model Inference**: Each model generates independent prediction | |
| 3. **Ensemble**: Performance-weighted average of predictions | |
| 4. **Uncertainty**: Confidence intervals from model disagreement | |
| ## 📈 Supported Assets | |
| - **US Stocks**: All NYSE, NASDAQ listed companies | |
| - **International Stocks**: Major global exchanges | |
| - **ETFs**: Index funds, sector ETFs, commodity ETFs | |
| - **Cryptocurrencies**: BTC-USD, ETH-USD, and major coins | |
| - **Market Indices**: S&P 500 (^GSPC), NASDAQ (^IXIC), Dow Jones | |
| ### Example Symbols | |
| ``` | |
| AAPL, GOOGL, MSFT, TSLA, NVDA, AMZN, META | |
| SPY, QQQ, VTI, IWM, GLD, SLV | |
| BTC-USD, ETH-USD, ADA-USD | |
| ^GSPC, ^IXIC, ^DJI, ^VIX | |
| ``` | |
| ## 🔬 Technical Details | |
| ### Feature Engineering Pipeline | |
| ```python | |
| def extract_features(data): | |
| features = pd.DataFrame() | |
| # Price features | |
| features['returns'] = data['Close'].pct_change() | |
| features['log_returns'] = np.log(data['Close'] / data['Close'].shift(1)) | |
| # Technical indicators | |
| features['rsi'] = ta.momentum.rsi(data['Close']) | |
| features['macd'] = ta.trend.macd_diff(data['Close']) | |
| features['bb_position'] = calculate_bollinger_position(data['Close']) | |
| # Volume features | |
| features['volume_ratio'] = data['Volume'] / data['Volume'].rolling(20).mean() | |
| features['obv'] = ta.volume.on_balance_volume(data['Close'], data['Volume']) | |
| return features | |
| ``` | |
| ### Model Training | |
| ```python | |
| # Time series cross-validation | |
| tscv = TimeSeriesSplit(n_splits=5) | |
| for train_idx, val_idx in tscv.split(features): | |
| X_train = features.iloc[train_idx] | |
| y_train = targets.iloc[train_idx] | |
| model.fit(X_train, y_train) | |
| score = model.score(X_val, y_val) | |
| ``` | |
| ### Performance Metrics | |
| - **R² Score**: Coefficient of determination | |
| - **MAE**: Mean Absolute Error | |
| - **RMSE**: Root Mean Square Error | |
| - **Directional Accuracy**: Percentage of correct direction predictions | |
| ## 📊 Example Output | |
| ``` | |
| 🤖 FINANCIAL AI PREDICTOR - AAPL | |
| 📊 CURRENT STATUS | |
| • Current Price: $182.50 | |
| • Analysis Date: 2025-01-15 14:30:00 | |
| • Recommendation: 🟢 BUY | |
| 🎯 ENSEMBLE PREDICTION | |
| • Next Day Target: $184.20 | |
| • Expected Return: +0.93% | |
| • Prediction Confidence: 73.5% | |
| 🤖 MODEL PERFORMANCE | |
| • Random Forest: R² = 0.0847 ± 0.0234 | |
| • Gradient Boost: R² = 0.0756 ± 0.0198 | |
| • Ridge: R² = 0.0542 ± 0.0156 | |
| 🧠 TOP PREDICTIVE FEATURES | |
| • rsi: 0.0234 | |
| • bb_position: 0.0198 | |
| • volume_ratio: 0.0176 | |
| • macd: 0.0145 | |
| ``` | |
| ## 🎯 Best Practices | |
| ### For Users | |
| 1. **Diversification**: Never rely on a single prediction | |
| 2. **Risk Management**: Use appropriate position sizing | |
| 3. **Validation**: Cross-check with fundamental analysis | |
| 4. **Timeframes**: Consider multiple prediction horizons | |
| 5. **Market Conditions**: Be aware of regime changes | |
| ### For Developers | |
| 1. **Feature Engineering**: Domain knowledge is crucial | |
| 2. **Validation**: Always use time series aware validation | |
| 3. **Overfitting**: Monitor out-of-sample performance | |
| 4. **Data Quality**: Clean and validate input data | |
| 5. **Model Updates**: Retrain periodically | |
| ## ⚠️ Important Disclaimers | |
| ### 🎓 Educational Purpose | |
| This tool is designed for: | |
| - Learning about machine learning in finance | |
| - Understanding feature engineering techniques | |
| - Demonstrating ensemble methods | |
| - Exploring financial data analysis | |
| ### 📉 Risk Warnings | |
| - **Not Financial Advice**: Predictions are for educational purposes only | |
| - **Past Performance**: Does not guarantee future results | |
| - **Model Limitations**: Cannot predict black swan events | |
| - **Market Risk**: All investments carry risk of loss | |
| - **Professional Advice**: Consult qualified financial advisors | |
| ### 🔬 Technical Limitations | |
| - **Market Efficiency**: Predictable patterns may be arbitraged away | |
| - **Regime Changes**: Models may fail during market shifts | |
| - **Data Dependency**: Quality depends on input data | |
| - **Overfitting**: Historical patterns may not persist | |
| ## 🛠️ Configuration | |
| ### Environment Variables | |
| ```bash | |
| # Optional: Set data cache directory | |
| export YFINANCE_CACHE_DIR="/path/to/cache" | |
| # Optional: Set log level | |
| export LOG_LEVEL="INFO" | |
| ``` | |
| ### Custom Configuration | |
| ```python | |
| # Modify model parameters | |
| MODELS_CONFIG = { | |
| 'random_forest': { | |
| 'n_estimators': 200, | |
| 'max_depth': 15, | |
| 'min_samples_split': 5 | |
| }, | |
| 'gradient_boost': { | |
| 'n_estimators': 150, | |
| 'max_depth': 8, | |
| 'learning_rate': 0.1 | |
| } | |
| } | |
| # Modify feature engineering | |
| FEATURE_CONFIG = { | |
| 'ma_periods': [5, 10, 20, 50], | |
| 'momentum_periods': [5, 10, 20], | |
| 'volatility_windows': [5, 20, 60] | |
| } | |
| ``` | |
| ## 🧪 Testing | |
| ```bash | |
| # Run basic tests | |
| python -m pytest tests/ | |
| # Test specific components | |
| python -m pytest tests/test_features.py | |
| python -m pytest tests/test_models.py | |
| # Run with coverage | |
| pytest --cov=financial_ai_predictor tests/ | |
| ``` | |
| ## 📚 Dependencies | |
| ### Core Requirements | |
| - **Python**: 3.8+ | |
| - **NumPy**: Scientific computing | |
| - **Pandas**: Data manipulation | |
| - **Scikit-learn**: Machine learning models | |
| - **yfinance**: Financial data | |
| - **TA**: Technical analysis indicators | |
| - **Gradio**: Web interface | |
| - **Plotly**: Interactive visualizations | |
| ### Optional Enhancements | |
| - **SciPy**: Advanced statistical functions | |
| - **Statsmodels**: Econometric models | |
| - **Jupyter**: Interactive development | |
| ## 🤝 Contributing | |
| 1. **Fork** the repository | |
| 2. **Create** a feature branch (`git checkout -b feature/amazing-feature`) | |
| 3. **Commit** your changes (`git commit -m 'Add amazing feature'`) | |
| 4. **Push** to the branch (`git push origin feature/amazing-feature`) | |
| 5. **Open** a Pull Request | |
| ### Development Setup | |
| ```bash | |
| # Clone for development | |
| git clone https://github.com/yourusername/financial-ai-predictor.git | |
| cd financial-ai-predictor | |
| # Install development dependencies | |
| pip install -r requirements-dev.txt | |
| # Run in development mode | |
| python financial_ai_predictor.py --debug | |
| ``` | |
| ## 📝 License | |
| This project is licensed under the MIT License - see the [LICENSE](LICENSE) file for details. | |
| ## 🙏 Acknowledgments | |
| - **Yahoo Finance**: For providing free financial data | |
| - **TA-Lib**: Technical analysis library | |
| - **Scikit-learn**: Machine learning framework | |
| - **Gradio**: Easy-to-use ML interfaces | |
| - **Plotly**: Interactive visualization library | |
| ## 📞 Support | |
| - **Issues**: [GitHub Issues](https://github.com/yourusername/financial-ai-predictor/issues) | |
| - **Discussions**: [GitHub Discussions](https://github.com/yourusername/financial-ai-predictor/discussions) | |
| - **Email**: your.email@example.com | |
| ## 🔮 Roadmap | |
| ### Version 2.0 | |
| - [ ] Deep learning models (LSTM, Transformer) | |
| - [ ] Alternative data integration | |
| - [ ] Multi-asset portfolio optimization | |
| - [ ] Real-time streaming predictions | |
| - [ ] Advanced risk metrics | |
| ### Version 1.5 | |
| - [ ] Options pricing models | |
| - [ ] Sector rotation analysis | |
| - [ ] Economic indicators integration | |
| - [ ] Backtesting framework | |
| - [ ] Performance attribution | |
| --- | |
| **⭐ Star this repository if you find it useful!** | |
| **🐛 Found a bug? Please open an issue.** | |
| **💡 Have an idea? Start a discussion.** | |
| --- | |
| *Built with ❤️ for the financial ML community* |