Bayesian Statistical Modeling with Stan, R, and Python

This book provides a practical introduction to Bayesian statistical modeling with Stan, covering theoretical background, workflow, probability distributions, nonlinear models, hierarchical models, and advanced topics. It offers examples and explanations for key concepts, with code and data on GitHub.

Format: Hardback
Length: 385 pages
Publication date: 25 January 2023
Publisher: Springer Verlag, Singapore

This comprehensive guide offers a thorough introduction to Bayesian statistical modeling using the widely popular probabilistic programming language Stan. Divided into four parts, the book begins by reviewing the theoretical foundations of modeling and Bayesian inference, presenting a streamlined workflow that demystifies the process and makes it more akin to engineering than art. The second part delves into the practical application of Stan,CmdStanR,and CmdStanPy, guiding readers from the very basics of regression analyses to advanced topics such as probability distributions, nonlinear models, hierarchical (multilevel) models, censoring, outliers, missing data, speed-up, parameter constraints, and convergence of MCMC. The third part explores a wide range of frequently employed modeling techniques, including censoring, outliers, missing data, speed-up, and parameter constraints, while discussing strategies to achieve convergence of MCMC. Finally, the fourth part delves into advanced real-world data topics, including longitudinal data analysis, state space models, spatial data analysis, Gaussian processes, Bayesian optimization, dimensionality reduction, model selection, and information criteria. Through numerous practical examples and clear explanations, the book demonstrates the versatility of Stan in solving a wide array of statistical modeling problems, with solutions often requiring just 30 lines of code. Written in a user-friendly style, the book assumes no prior knowledge of statistical modeling and can be easily adapted to various fields. It provides full explanations of code and math formulas, enabling readers to extend models for their specific problems. All the code and data used in the book are available on GitHub, making it an invaluable resource for anyone interested in learning Bayesian statistical modeling.

Weight: 840g
Dimension: 160 x 243 x 27 (mm)
ISBN-13: 9789811947544
Edition number: 1st ed. 2022