This report explores key tree-based statistical learning methods including regression trees, bagging, random forests, and classification trees. The analysis focuses on model interpretation, variable importance, and prediction accuracy. Using the Carseats and OJ datasets from the ISLR2 package, both regression and classification tasks are examined. The report emphasizes cross-validation for model selection, the impact of pruning on performance, and comparisons between pruned and unpruned trees. Results highlight the trade-offs between model complexity and generalization in both regression and classification settings.