A machine learning project to forecast Google's stock price using historical data and multiple regression models.

This project:

• Loads historical stock data (Google - GOOG)
• Performs feature engineering (volatility, percentage change)
• Trains and evaluates three models:
  • Support Vector Machine (SVM)
  • Linear Regression
  • Random Forest Regression
• Forecasts future stock prices
• Visualizes predictions vs actual prices

• ✅ Support Vector Machine (SVR)
• ✅ Linear Regression
• ✅ Random Forest Regressor (100 trees)

Each model is saved using pickle for reuse.

Each model's accuracy is evaluated using .score() on a test set.

You’ll see results like:

Model Accuracy (SVM): 0.79
Model Accuracy (Linear Regression): 0.83
Model Accuracy (Random Forest): 0.88

(Note: actual numbers may vary depending on data range.)

• Actual stock price from 2004 onward
• Forecasts shown from the most recent date forward
• Colors:
  • 🔵 Actual Price
  • 🟢 SVM
  • 🔴 Linear Regression
  • 🟠 Random Forest

├── GOOG.csv                   # Input dataset (downloaded)
├── stock_forecasting.py       # Full training and visualization script
├── svm_linear.pickle          # Saved SVM model
├── linearregression.pickle    # Saved Linear Regression model
├── randomforest.pickle        # Saved Random Forest model
├── forecast_plot.png          # Result image (optional)
├── requirements.txt           # Python dependencies
└── README.md                  # Project documentation

# Clone the repo
git clone https://github.com/yourusername/your-repo.git
cd your-repo

# Install requirements
pip install -r requirements.txt

# Run the model
python stock_forecasting.py

pandas
numpy
matplotlib
scikit-learn
pickle-mixin

Mohammed Asaad
📧 mo.asaad999@gmail.com
🌍 LinkedIn

This project is licensed under the MIT License.
Feel free to use, modify, or build on it with attribution.