📈 Gold Price Prediction

Predicting Gold (GLD) ETF Prices Using Machine Learning

---

📖 Overview

This project builds a Random Forest Regression model to predict the price of the SPDR Gold Shares (GLD) ETF using historical financial market data. The model leverages correlated financial instruments — including the S&P 500 index, crude oil (USO), silver (SLV), and the EUR/USD exchange rate — as predictive features to estimate gold price movements.

The complete analysis pipeline — from data loading and exploratory analysis through model training and evaluation — is implemented in a single, reproducible Jupyter Notebook designed for Google Colab.

---

✨ Key Features

• 📊 Exploratory Data Analysis — Statistical summaries, null-value checks, distribution plots, and correlation heatmaps to deeply understand the dataset before modeling.
• 🤖 Random Forest Regressor — Ensemble learning model with 100 decision trees for robust, high-accuracy price predictions.
• 📈 Visual Model Evaluation — Side-by-side comparison of actual vs. predicted gold prices using line plots for intuitive performance assessment.
• 📐 Quantitative Metrics — Model accuracy measured using R-squared error and Mean Absolute Error (MAE).
• ☁️ Cloud-Ready — Pre-configured for seamless execution on Google Colab with zero local setup required.

---

🗂️ Project Structure

Gold_Price_Prediction-main/
│
├── Gold_Price_Prediction.ipynb   # Main Jupyter Notebook (full pipeline)
├── gold_price_data.csv           # Dataset (2,290 trading days)
├── Gold Price Prediction.docx    # Project documentation
├── README.md                     # This file
└── LICENSE                       # MIT License

---

📋 Dataset

The dataset contains 2,290 daily observations spanning from January 2, 2008 to May 16, 2018, with no missing values.

Column	Description	Type
Date	Trading date	Object
SPX	S&P 500 Index closing price	Float64
GLD	SPDR Gold Shares ETF price (Target)	Float64
USO	United States Oil Fund price	Float64
SLV	iShares Silver Trust price	Float64
EUR/USD	EUR to USD exchange rate	Float64

Key Statistics

Metric	SPX	GLD	USO	SLV	EUR/USD
Mean	1,654.32	122.73	31.84	20.08	1.2837
Std	519.11	23.28	19.52	7.09	0.1315
Min	676.53	70.00	7.96	8.85	1.0390
Max	2,872.87	184.59	117.48	47.26	1.5988

---

🔬 Methodology

1. Data Collection & Processing

• Load the CSV dataset into a Pandas DataFrame.
• Inspect for missing values, data types, and basic statistical measures.

2. Exploratory Data Analysis

• Distribution Analysis — Visualize the GLD price distribution using Seaborn's displot.
• Correlation Analysis — Compute and visualize feature correlations using a heatmap to identify relationships between financial instruments.

3. Feature Engineering

• Features (X): SPX, USO, SLV, EUR/USD
• Target (Y): GLD
• Drop the Date column (non-numeric) and the target variable from the feature set.

4. Train-Test Split

• 80/20 split with random_state=2 for reproducibility.
• Training set: 1,832 samples
• Test set: 458 samples

5. Model Training

• Algorithm: RandomForestRegressor from Scikit-learn
• Hyperparameters: n_estimators=100 (100 decision trees in the ensemble)

6. Model Evaluation

• Predictions generated on the held-out test set.
• Performance assessed using:
  • R-squared (R²) — Measures the proportion of variance explained by the model.
  • Mean Absolute Error (MAE) — Average absolute difference between predicted and actual values.
• Visualization: Actual vs. predicted prices plotted for visual comparison.

---

🛠️ Technology Stack

Category	Libraries
Data Manipulation	NumPy, Pandas
Visualization	Matplotlib, Seaborn
Machine Learning	Scikit-learn (RandomForestRegressor, metrics)
Environment	Jupyter Notebook, Google Colab

---

🚀 Getting Started

Option 1: Google Colab (Recommended)

The quickest way to run this project — no local installation needed:

1. Click the "Open in Colab" badge at the top of this README.
2. Upload gold_price_data.csv to the Colab runtime when prompted (or modify the path).
3. Run all cells sequentially (Runtime → Run all).

Option 2: Run Locally

Prerequisites

• Python 3.9 or higher
• pip (Python package manager)

Steps

# 1. Clone the repository
git clone https://github.com/MYoussef885/Gold_Price_Prediction.git
cd Gold_Price_Prediction

# 2. Create and activate a virtual environment
python -m venv venv

# Windows
venv\Scripts\activate

# macOS / Linux
source venv/bin/activate

# 3. Install dependencies
pip install numpy pandas matplotlib seaborn scikit-learn jupyter

# 4. Launch Jupyter Notebook
jupyter notebook Gold_Price_Prediction.ipynb

---

📈 Results

The Random Forest Regressor demonstrates strong predictive performance on the test set:

• ✅ High R² score — The model explains a significant proportion of variance in gold prices.
• ✅ Low MAE — Predicted values closely track actual GLD prices.
• ✅ Visual alignment — The actual vs. predicted price plot shows tight correspondence, particularly for mid-range gold prices.

Note: The model performs best when market conditions follow historical patterns observed in the 2008–2018 training period. Performance may vary for out-of-distribution data.

---

🔮 Future Improvements

• Add time-series features (moving averages, RSI, MACD)
• Experiment with additional models (XGBoost, LSTM, Gradient Boosting)
• Implement hyperparameter tuning via GridSearchCV or RandomizedSearchCV
• Incorporate additional macroeconomic indicators (interest rates, inflation)
• Build an interactive dashboard for real-time predictions
• Extend the dataset to include more recent trading data

---

🤝 Contributing

Contributions, issues, and feature requests are welcome! Feel free to:

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

---

📝 License

This project is licensed under the MIT License — see the LICENSE file for details.

---

🙏 Acknowledgements

• MYoussef885 — Original project author
• Siddhardhan — Project mentor and tutorial guidance
• The open-source community behind NumPy, Pandas, Scikit-learn, Matplotlib, and Seaborn

---

_{Built with ❤️ for data science and financial analysis}