Logistic regression is one of the most widely used tools in statistical modelling, yet the gap between textbook theory and real-world practice remains a persistent challenge for students, researchers, and practitioners alike. This book bridges that gap with a comprehensive, hands-on guide to binary outcome modelling that goes well beyond the basics.
Built on a foundation of rigorous statistical theory, the book tackles the messy realities that applied analysts routinely face, including separation, rare events bias, overdispersion, and multicollinearity, offering clear and practical strategies for each. Rather than treating these as edge cases, the author positions them as central concerns deserving serious methodological attention. Modern variable selection strategies are examined in depth, contrasting traditional approaches with contemporary regularisation methods, while advanced topics such as Bayesian logistic regression and propensity score methods broaden the reader's analytical toolkit.
Throughout, statistical theory is integrated with computational methods and domain knowledge, grounded in reproducible R code, simulated examples, and real-world applications drawn from fields where the stakes of getting it wrong are high.
Whether you are an advanced undergraduate or graduate student studying regression modelling or applied statistics, a researcher navigating imbalanced outcomes in epidemiology or finance, or a data scientist seeking reliable methods for classification problems, this book offers the depth and practicality to meet you where you are and take your work further.
Table of Contents:
1.Review of Logistic Regression Fundamentals 2. Complete and Quasi-Complete Separation 3. Rare Events Bias 4. Overdispersion and Multiple Link Functions 5. Variable Selection Methods 6. Multicollinearity 7. Non-linearity in Predictors 8: Interaction Effects 9. Model Diagnostics and Goodness-of-Fit 10: Model Validation and Prediction 11. Logistic Regression for Longitudinal Data 12. Multinomial and Ordinal Logistic Regression 13. Handling Missing Data 14. Survey Data and Complex Sampling 15. Bayesian and Causal Methods 16. Reporting Standards and Domain Applications. Hints and Selected Solutions. Solutions to Exercises
About the Author :
Hassan Doosti is Program Director of the Master of Data Science and Senior Lecturer in Statistics in the School of Mathematical and Physical Sciences at Macquarie University, Sydney, Australia. He is the author and editor of four books, including Nonparametric Flexible Curve Estimation (Springer Nature, 2024), Ethics in Statistics: Opportunities and Challenges (Ethics International Press, 2024), Practical Biostatistics for Medical and Health Sciences (Springer Nature, 2024), co-authored with Hassan Saneii, and Long Memory Time Series Analysis (Chapman and Hall/CRC, 2026), co-authored with Gnanadarsha Sanjaya Dissanayake.
