• Predicted next day returns for AAPL, AMZN, GOOGL, MSFT, and NVDA using XGBoost, Bagging, Random Forest, SVM, Lasso, and Linear Modeling, achieving greatest accuracy of 55.96%, a regression problem focusing on feature engineering
• Performed data cleaning techniques like imputation, tuned hyperparameters, chose best learning rate, extracted variable importance using SHAP values, generated confusion matrices, and plotted actual v. predicted returns using plotly
Please see the final report for more details and the final presentation to learn more.