The objective of this assignment is to replicate the instructor’s visualizations accurately. Viz 1 uses plotly while Viz 2-4 use highcharter. Each task is designed to optimize certain features of the plot, intentionally leaving some aesthetics imperfect to highlight specific functionalities.

These exercises involve classification tasks using support vector machines and tuning of hyper parameters to improve model performance.

These exercises involve classification and regression tasks utilizing single decision trees, random forest, and BART (Bayesian Additive Regression Trees) models. Cross validation is employed to optimize the pruning of decision trees for improved performance.

These exercises involve regression techniques, including polynomial regression, step functions, and generalized additive models (GAMs), balancing flexibility with interpretability.

These exercises involve selection and regularization algorithms such as lasso, ridge, and subset selection. These techniques are used to choose features, prevent overfitting and account for multicollinearity, with the overall goal to improve prediction accuracy of linear regression models.

These exercises involve predictive modeling with focus on resampling methods like the validation set approach, cross-validation, and bootstrapping.

These exercises involve logistic regression, linear discriminant analysis (LDA), quadratic discriminant analysis (QDA), naive Bayes, and k-nearest neighbors (KNN) for classification tasks. These methods were applied to predict various outcomes based on different sets of predictors.

These exercises involve multiple linear regression analysis, including data exploration, model fitting, diagnostic assessments, interpretation of coefficients, hypothesis testing and identification of influential points.