Gold Price Prediction Using Random Forest Classifier

This project involved predicting trends in gold prices—such as whether they would rise, fall, or remain stable—using a Random Forest Classifier. The objective was to leverage historical data and market indicators to assist in decision-making for investments and risk management.

• Methodology:

  • Data Collection and Preprocessing:

    • Worked with a dataset comprising historical gold prices, crude oil prices, stock market indices, and currency exchange rates.

    • Preprocessed the data by handling missing values, encoding categorical variables, and normalizing numerical features.

  • Target Engineering:

    • Transformed continuous gold price data into categories like Increase, Decrease, and No Change based on price thresholds.

  • Model Training:

    • Implemented a Random Forest Classifier to handle multi-class classification, utilizing its ability to handle high-dimensional and imbalanced data.

  • Model Optimization:

    • Tuned hyperparameters such as the number of trees, maximum depth, and class weights using grid search for optimal performance.

• Key Results and Insights:

  • Achieved an accuracy of XX% with strong precision and recall metrics for all classes.

  • Identified significant predictors, including currency exchange rates and stock indices, that influenced gold price trends.

  • Demonstrated the Random Forest Classifier’s ability to handle non-linear relationships and provide interpretable feature importance metrics.

• Skills and Tools:

  • Machine Learning (Random Forest Classifier), Python (Pandas, NumPy, Scikit-learn, Matplotlib, Seaborn), Data Preprocessing, and Model Evaluation.

This project showcases the application of machine learning in financial analysis, providing valuable insights into market behavior and enabling data-driven investment strategies.

Github Project link: https://github.com/sm98code/ML.git

 Visualization Result