Applied Statistics with Python, Volume II: Multivariate Models

Applied Statistics with Python, Volume II: Multivariate Models is a focused, code-integrated statistics text for readers who want to move beyond introductory methods and into the practical middle ground where modeling, prediction, and interpretation meet. Leon Kaganovskiy uses Python as part of the learning process rather than as a detached add-on, making the examples feel anchored in the language many data scientists and analysts actually use.

From ANOVA to predictive modeling

The book begins with analysis of variance and then moves through a broad toolkit of multivariate methods: multiple regression, polynomial regression, interaction terms, model selection, shrinkage methods, logistic regression, K-nearest neighbors, support vector classification, nonlinear models, tree-based methods, ensembles, principal components, and clustering. The result is a structured path from classical inference toward modern statistical learning.

Python woven into every step 💻

One of the book’s defining features is its emphasis on working code. Rather than treating software as a separate lab exercise, the text integrates Python and the sklearn ecosystem directly into the explanation of the statistics. That approach should appeal to readers who learn best by seeing methods implemented, not just described.

Who this volume is for

The preface identifies a broad academic audience: undergraduate and first-year graduate students in business, economics, biology, social sciences, natural sciences, and related fields. The book also appears suitable for more advanced readers who want a compact, practice-oriented refresher on core multivariate methods in Python.

A practical companion to modern statistics

Because the chapters progress from familiar inferential ideas into machine-learning-oriented techniques, the volume works well as both a course text and a self-study reference. Readers who want a single resource for applied statistical modeling in Python will find a coherent, classroom-tested presentation here.