This book provides a detailed introductory survey of multivariate statistical modeling based on generalized linear models, utilizing real data from diverse fields. The authors present technical details and proofs in an appendix to ensure accessibility for non-experts, serving as a reference or tutorial for concepts like the EM algorithm, numerical integration, and MCMC. The revised edition extensively incorporates growing Bayesian concepts, a completely rewritten chapter on nonparametric and semiparametric generalized regression, and expanded coverage of random effects models with Bayesian approaches. It also includes state-space and hidden Markov models with extensions for spatial and spatiotemporal data, aiming to connect theoretical ideas with practical data applications.

This book is designed for applied statisticians, graduate students of statistics, and researchers in econometrics, biometrics, and the social sciences who have a strong interest in statistics and data analysis. The authors' approach of relating underlying theoretical ideas directly to the data at hand makes it particularly suitable for those seeking practical application of statistical methods. The inclusion of advanced topics and Bayesian concepts in the new edition reflects the evolving landscape of statistical modeling. Its comprehensive nature and focus on real-world examples position it as a valuable resource for both learning and reference.