Prepare
Due: September 17, 2024
prepare
Learning objectives
- Identify 1-to-1 and 1-to-many transformation methods for encoding numeric predictors
- Implement transformations for numeric predictors to reduce skewness
- Introduce basis expansions and splines to incorporate non-linear relationships in predictors
- Define Generalized Additive Model (GAM) and multivariate adaptive regression spline (MARS) for modeling non-linear relationships
- Model nonlinear relationships between predictors and the response variable
Preparations
📖 Read
- Feature engineering and selection (FES) - ch 6.1-.2 by Max Kuhn and Kjell Johnson