Linear regression with a multiple predictors II

Lecture 17

2024-06-10

Announcements

  • June 12 is the last day to withdraw from the class.
  • If you are struggling, please attend office hours and ask questions on Ed!
  • Feel free to ask questions regarding Exam 1 on Ed.

Lab 4 recap- interpreting log() response

Goal: Interpret the coefficients based on the following regression:

\[ \log(Y) = \beta_0 + \beta_1 x_1 + \beta_2 x_2 +\cdots. \] We don’t want our interpretation based on the \(\log()\) scale:

\[ e^{\log(Y)} = Y = e^{ \beta_0} e^{ \beta_1 x_1}e^{ \beta_2 x_2 }e^{ \cdots}. \] So, say we want to interpret \(\beta_2\) where \(x_2\) is a categorical group membership variable: \[ Y = e^{ \beta_0} e^{ \beta_1 x_1}\color{red}{e^{ \beta_2 \times 1 }}e^{ \cdots}. \]

Application exercise: ae-12

  • Let’s finish AE-12!

Model selection and overfitting

Goals

  • Review prediction and interpretation of model results
  • Review main and interaction effects models
  • Discuss model selection further

Application exercise: ae-13-modeling-loans

  • Go to your project called ae.
  • If there are any uncommitted files, commit them, and push.
  • Work on ae-13-modeling-loans.qmd.