This report details the exploratory data analysis and modeling of the Boston housing dataset, focusing on predicting median home values using linear regression and classification and regression trees (CART). The analysis involved data splitting, exploratory analysis, model fitting through various regression techniques, and cross-validation, revealing that regression trees performed best in terms of mean squared error across trials. Additionally, a comparison of logistic regression and CART models highlighted the effectiveness of the CART model based on misclassification rates.