Leave Pima Indians alone: binary regression as a benchmark for Bayesian\n computation

2015/06/29 by ChopinNicolas, Chopin, Nicolas, James Ridgway +1 · 3 citations
Computer Science · Mathematics · #Bayesian Methods and Mixture Models #Computation (stat.CO) #FOS: Computer and information sciences #Markov Chains and Monte Carlo Methods #Methodology (stat.ME) #Statistical Methods and Bayesian Inference

paper · pdf · doi:10.48550/arxiv.1506.08640

openalex publication_date 2015/06/29 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Abstract. Whenever a new approach to perform Bayesian computation is\nintroduced, a common practice is to showcase this approach on a binary\nregression model and datasets of moderate size. This paper discusses to which\nextent this practice is sound. It also reviews the current state of the art of\nBayesian computation, using binary regression as a running example. Both\nsampling-based algorithms (importance sampling, MCMC and SMC) and fast\napproximations (Laplace and EP) are covered. Extensive numerical results are\nprovided, some of which might go against conventional wisdom regarding the\neffectiveness of certain algorithms. Implications for other problems (variable\nselection) and other models are also discussed.\n

Citations

• Annealed importance sampling
• Bias reduction of maximum likelihood estimates
• Bayesian Inference for Logistic Models Using Pólya–Gamma Latent Variables
• Bayesian Analysis of Binary and Polychotomous Response Data

Cited by

• Approximate Bayesian inference for cumulative probit regression models
• Mixing times of data-augmentation Gibbs samplers for high-dimensional probit regression
• A Class of Conjugate Priors for Multinomial Probit Models which Includes\n the Multivariate Normal One

Related

• On computational tools for Bayesian data analysis
• Bayesian computational methods
• Approximate Methods for Bayesian Computation
• Bayesian computation: a perspective on the current state, and sampling backwards and forwards
• Bayesian analysis of dynamic binary data: A simulation study and application to economic index SP