Bayesian Statistics, New Generations New Approaches: BAYSM 2022, Montréal, Canada, June 22–23 (Springer Proceedings in Mathematics & Statistics, 435)

This book hosts the results presented at the 6th Bayesian Young Statisticians Meeting 2022 in Montréal, Canada, held on June 22–23, titled "Bayesian Statistics, New Generations New Approaches". This collection features selected peer-reviewed contributions that showcase the vibrant and diverse research presented at meeting. This book is intended for a broad audience interested in statistics and aims at providing stimulating contributions to theoretical, methodological, and computational aspects of Bayesian statistics. The contributions highlight various topics in Bayesian statistics, presenting promising methodological approaches to address critical challenges across diverse applications. This compilation stands as a testament to the talent and potential within the j-ISBA community. This book is meant to serve as a catalyst for continued advancements in Bayesian methodology and its applications and encourages fruitful collaborations that push the boundaries of statistical research. Preface Contents Bayesian Emulation of Complex Computer Models with Structured Partial Discontinuities 1 Introduction 2 Bayesian Emulation with Partial Discontinuities 2.1 Emulation of Computer Models 2.2 Torn Embeddings in Higher Dimensions 2.3 Controlling for the Induced Local Warping Effect 2.4 Controlling for the Global Impact of the Embedding Using Non-stationary Emulation 3 Application: TNO OLYMPUS Well Placement Optimisation Challenge 4 Conclusion References A Variational Bayes Approach to Factor Analysis 1 Background 2 Methods 3 Results 4 Conclusions References Scalable Model Selection for Staged Trees: Mean-posterior Clustering and Binary Trees 1 Introduction 2 Preliminaries 2.1 Staged Trees 2.2 Conjugate Learning and Model Selection 3 Methods 3.1 Totally Ordered Hyperstage 4 Mean Posterior Probabilities 4.1 Resize Operator 5 A Comparative Analysis of Competing Methodologies 6 Christchurch Health and Development Study Example 7 Discussion References Speeding up the Zig-Zag Process 1 Introduction 2 The SUZZ Process 3 Theoretical Results 4 Numerical Examples References Extended Stochastic Block Model with Spatial Covariates for Weighted Brain Networks 1 Introduction 2 Poisson Extended Stochastic Block Model 2.1 Cohesion Function 2.2 Posterior Inference 3 Application to Brain Networks 3.1 Uncertainty Quantification 4 Discussion References Approximate Bayesian Inference for Smoking Habit Dynamics in Tuscany 1 Introduction 2 Smoking Habit Compartmental Model 2.1 The Probabilistic Model 2.2 Approximate Bayesian Computation for the SHC Model 3 Results 4 Discussion References Mixing Times of a Gibbs Sampler for Probit Hierarchical Models 1 Introduction 2 Probit Hierarchical Models 3 Theoretical Results on Mixing Times 4 Numerical Illustration 5 Conclusions References A Note on the Dependence Structure of Hierarchical Completely Random Measures 1 Introduction 2 Hierarchical Completely Random Measures 3 Dependence Structure 4 Discussion 5 Proofs References Observed Patterns of Heat Wave Intensities with Respect to Time and Global Surface Temperature 1 Introduction 2 Methods 3 Applications 3.1 Heat Wave Maximum Intensity Over Time 3.2 Heat Wave Maximum Intensity and Global Surface Temperature 4 Conclusions References Expectation Propagation for the Smoothing Distribution in Dynamic Probit 1 Introduction 2 Literature Review 3 Expectation Propagation (EP) for the Dynamic Probit 3.1 Implementation Without p n times p npntimespn Matrix Inversions 3.2 Implementation Without p n times p npntimespn Matrix Updates 3.3 Computational Costs 4 Financial Illustration 5 Discussion References