Data Science Methods

Welcome!

This book presents an introductory course in data science with a focus on the practical aspects of doing data analysis and building prediction models. The book begins with exploratory data analysis and focuses on the fundamentals of navigating a dataset, specifying a question, exploring the characteristics of individual variables, and describing the relationships between variables. The second part introduces linear models for prediction along with the training and evaluation of core machine learning algorithms. A novel contribution of this book is the introduction of the statistical approach to causal inference. Topics such as the potential outcomes framework and confounding are covered, as well as the use of matching methods for estimating average treatment effects. Finally, the book covers issues that arise when doing data analysis in the real world through a series of case studies that illustrate some of those issues and how to address them.