An EPAT project compares Logistic Regression, Random Forest, and XGBoost on the S&P 500 and Brent Crude, testing machine learning trading signals against buy and hold from 2022 to 2025.