Gold Price Prediction Using Machine Learning

This project demonstrates the prediction of gold prices using a machine learning model (Random Forest Regressor). The model is trained on historical gold price data to predict future values with a focus on correlation analysis and error measurement (R squared).

Dataset

The dataset used contains historical data of gold prices and related financial parameters, loaded from a CSV file. The date approximation is of around 10 years of data.

Libraries Used

• pandas: Data manipulation and analysis
• numpy: Numerical computations
• matplotlib: Data visualization
• seaborn: Advanced data visualization
• scikit-learn: Machine learning models and metrics

Steps Involved

1. Data Loading: The dataset is loaded using pandas from a CSV file.
2. Data Cleaning: Checked for missing values and converted columns to numeric types.
3. Exploratory Data Analysis (EDA):
   • Used correlation matrix to find relationships between variables.
   • Visualized the distribution of gold prices.
4. Model Training:
   • Features (X) were extracted by removing the Date and GLD columns.
   • The target (Y) is the GLD (gold price).
   • The data is split into training and testing sets using an 80/20 split.
5. Prediction:
   • The Random Forest Regressor model is trained on the training set.
   • The model predicts the values on the test set, and the performance is evaluated using the R-squared error.
6. Visualization:
   • Comparison of actual vs predicted gold prices using a line plot.

Model Performance

• The model's accuracy is measured using the R-squared error, where a higher score indicates better performance.

Visualizations

• Correlation Heatmap: Shows the relationships between different financial factors and gold prices.
• Actual vs Predicted Plot: Compares the model's predictions with actual values.

Requirements

• Python 3.9.6 or above
• Libraries: pandas, numpy, matplotlib, seaborn, scikit-learn