Linear modeling
Objectives
By the end, you’ll be able to:
- Fit robust predictive models for both regression and classification tasks.
- Judge feature contributions via confidence intervals, t‑tests, and F‑tests.
- Select the best subset of predictors for optimal performance.
Concepts Covered
- Simple vs. Multiple Linear Regression
- Key Assumptions (Gauss–Markov Theorem)
- Inference & Statistical Testing
- t‑tests, F‑tests, Confidence Intervals
- Residual & Diagnostic Analysis
- Categorical Variables
- Dummy Encoding, Polynomial Regression
- Generalized Linear Models
- Logistic Regression, Link Functions