Reasoning Through Regression | Bahadır Yüzbaşı

Inference, Prediction, and Regularization with R

Author: Prof. Dr. Bahadır Yüzbaşı

This companion connects the book’s mathematics with runnable R applications. Work through the chapters in order: define the question, understand the mathematical object, run the code, inspect the result, and explain the conclusion.

Companion edition: 27 September 2026. The worked examples connect theory with reproducible calculations. Chapter 10 includes the verified revision using observation-weighted validation scores and training-only penalty grids. The textbook remains an author manuscript; this site publishes its chapter application code.

Chapter	Topic	Application focus
1	Questions, Data, and Modeling Goals	Mean and squared loss
2	Simple Linear Regression	Fit a line and explain its residuals
3	Multiple Regression and Conditional Interpretation	Interpret an adjusted comparison
4	Uncertainty, Inference, and Prediction Intervals	Match uncertainty to its target
5	Model Checking and Assumption Failures	Ask distinct diagnostic questions
6	Categorical Predictors, Interactions, and Transformations	Understand model representation
7	Multicollinearity and Unstable Estimates	Separate agreement from accuracy
8	Generalization, Resampling, and Model Selection	Define and pool prediction loss
9	Bias, Variance, and Ridge Regression	Follow directional ridge shrinkage
10	Lasso, Elastic Net, and Selection Stability	Explain sparsity and selection
11	Logistic Regression and Probability Predictions	Evaluate probabilities
12	A Complete Regression Study	Connect a complete study

The chapter pages give prerequisites, input provenance and a short execution command. Successful execution is one part of learning; retain your initial explanation and return later to an unseen check.

The downloadable package contains student application code and execution guidance. It does not include instructor answer keys.