Statistical Learning from a
Regression Perspective

This is a very nice book that presents much of viewing machine learning as a regression problem. These are some notes and solutions that I came up with as I worked some of the problems therein.

Readers unfamiliar with this book can see what others have said here.

• Here
  

• Chapter 1 (Statistical Learning as a Regression Problem):
  • chap_1_prob_1.R (the airquality dataset)
    
  • chap_1_prob_2.R (complexity of the fitting function)
    
  • chap_1_prob_3.R (overfitting the data)
    
• Chapter 2 (Regression Splines and Regression Smoothers):
  • chap_2_prob_1.R (smoothers with one predictor)
    
  • chap_2_prob_2.R (smoothers with two predictors)
    
  • chap_2_prob_3.R (smoothers with more than two predictor)
    
  • chap_2_prob_4.R (smoothers with a binary response variable)
    
• Chapter 3 (Classification and Regression Trees (CART)):
  • chap_3_prob_1.R (CART vs. linear models)
    
  • chap_3_prob_2.R (classification with CART)
    
  • chap_3_prob_3.R (regression with CART)
    
• Chapter 4 (Bagging):
  • chap_4_prob_1.R (linear regression, CART, and bagging)
    
  • chap_4_prob_2.R (the Freedman dataset)
    
  • chap_4_prob_3.R (predicting frogs)
    
• Chapter 5 (Random Forests):
  • chap_5_prob_1.R (nonlinear curve fitting with random forests)
    
  • chap_5_prob_2.R (various options of the randomForest code)
    
  • chap_5_prob_3.R (classifying diabetes with random forests)
    
• Chapter 6 (Boosting):
  • chap_6_prob_1.R (nonlinear curve fitting with gradient boosting)
    
  • chap_6_prob_2.R (gradient boosting on the Leinhardt dataset)
    
• Chapter 7 (Support Vector Machines):
  • chap_7_prob_1.R (SVMs vs. logistic regression)
    
  • chap_7_prob_2.R (SVMs on the Pima.tr dataset)