Main points of chapter 2, introducing statistical learning

• Statistical learning: want to estimate a function.
  
• Given an \(x\), we want to predict its \(y\);
• we want to infer how \(x\) relates to \(y\);
• we want to evaluate the quality of our predictions and inferences.
• Training vs test error.
  
• Overfitting and model flexibility.
  
• error = variance + bias + irreducible error.
  
• Often a trade off between bias and variance. What does this say about model flexibility?