Handbook of Bayesian Variable Selection

Bayesian variable selection has evolved significantly in the past 30 years, with large data sets enabling simpler interpretation, avoiding overfitting, and providing insights into mechanisms. The Handbook of Bayesian Variable Selection provides a comprehensive review of theoretical, methodological, and computational aspects, targeting graduate students and established researchers.

Format: Paperback / softback
Length: 490 pages
Publication date: 29 January 2024
Publisher: Taylor & Francis Ltd

Over the past three decades, Bayesian variable selection has undergone significant advancements, fueled by the availability of vast data sets. The process of identifying relevant variables to include in a model not only simplifies interpretation, prevents overfitting and multicollinearity, but also offers valuable insights into the underlying mechanisms of observed phenomena. In scenarios where the number of potential predictors surpasses the sample size, variable selection becomes particularly crucial, as sparsity can be reasonably assumed.

The Handbook of Bayesian Variable Selection serves as a comprehensive resource that delves into the theoretical, methodological, and computational aspects of Bayesian methods for variable selection. It encompasses a wide range of topics, including spike-and-slab priors, continuous shrinkage priors, Bayes factors, Bayesian model averaging, partitioning methods, and variable selection in decision trees and edge selection in graphical models. Designed with graduate students and established researchers in mind, this handbook provides a thorough understanding of the latest developments in the field. It also serves as a valuable reference for individuals interested in applying existing methods or exploring methodological extensions.

The handbook is organized into four parts:

Spike-and-Slab Priors: This section explores the use of spike-and-slab priors, which are characterized by a combination of a point mass and a continuous distribution. It discusses their theoretical foundations, computational considerations, and applications in various modeling scenarios.

Continuous Shrinkage Priors: This section introduces continuous shrinkage priors, which are characterized by a smooth, non-parametric distribution that allows for flexible modeling of complex relationships between variables. It discusses their theoretical underpinnings, implementation in R, and their applications in various domains, such as survival analysis and machine learning.

Extensions to Various Modeling: This section explores the extension of Bayesian variable selection methods to various modeling frameworks, including generalized linear models, mixed models, and spatial models. It discusses the challenges and opportunities associated with these extensions and provides practical examples using R.

Other Approaches to Bayesian Variable Selection: This section covers other approaches to Bayesian variable selection, such as information-theoretic approaches, decision-theoretic approaches, and hierarchical models. It discusses their theoretical foundations, computational considerations, and applications in different domains.

In addition to its comprehensive coverage, the handbook offers valuable features:

Comprehensive Review: The handbook provides a comprehensive review of the theoretical foundations, methodological considerations, and computational aspects of Bayesian variable selection. It covers a wide range of topics, ensuring that readers have a thorough understanding of the field.

Divided into Four Parts: The handbook is divided into four parts, each focusing on a specific aspect of Bayesian variable selection. This organization makes it easy to navigate and find relevant information.

Theoretical and Methodological Aspects: The handbook covers both theoretical foundations and practical methodologies, providing readers with a balanced perspective on the field. It includes worked-out examples with R code provided in the online supplement, allowing readers to apply the discussed methods directly to their data.

Contributions by Experts: The handbook features contributions from experts in the field, ensuring that the content is up-to-date and authoritative.

Supplementary Material: The handbook includes a website with code, data, and other supplementary material, providing readers with additional resources to further explore the topics covered.

In conclusion, the Handbook of Bayesian Variable Selection is a valuable resource for graduate students, established researchers, and practitioners interested in understanding and applying Bayesian methods for variable selection. Its comprehensive coverage, theoretical and methodological insights, and practical examples make it an essential tool for anyone seeking to advance their knowledge in this field.

Weight: 910g
Dimension: 254 x 178 (mm)
ISBN-13: 9780367543785