Jonah Sol Gabry

Software and papers

Jump to:

• Stan
• R packages
• Papers
• Tutorials and educational materials

---

Stan

Stan is a statistical modeling language along with inference algorithms for full Bayesian inference, approximate Bayesian inference, and penalized maximum likelihood estimation. Stan is implemented in C++ but we provide interfaces for the command line, R and Python (and more).

• Stan website
• The Stan Forums (on Discourse)
• stan-dev GitHub organization
• Stan Development Team

R packages

rstan

The rstan package is the R interface to Stan.
website | CRAN | source code on GitHub

cmdstanr

The cmdstanr package is an alternative interface to Stan from R that is compatible with more recent Stan releases than rstan. It is not currently available on CRAN but can be installed easily from GitHub or R-universe.
website | source code on GitHub

rstanarm

The rstanarm package is for Bayesian applied regression modeling (ARM) via Stan. It is an appendage to the rstan package that enables some of the most common applied regression models to be estimated using Markov Chain Monte Carlo, variational approximations to the posterior distribution, or optimization. The rstanarm package allows these models to be specified using the customary R modeling syntax (e.g., like that of glm with a formula and a data.frame).
website | CRAN | source code on GitHub

bayesplot

The bayesplot R package provides a library of plotting functions for use after fitting Bayesian models (typically with MCMC).
website | CRAN | source code on GitHub

A paper about bayesplot and visualization in the Bayesian workflow more generally:

• Jonah Gabry, Daniel Simpson, Aki Vehtari, Michael Betancourt, and Andrew Gelman. (2019). Visualization in Bayesian workflow. J. R. Stat. Soc. A, 182: 389-402. doi:10.1111/rssa.12378. published version | arXiv preprint

loo

The loo R package is for efficient approximate leave-one-out cross-validation for Bayesian models.
website | CRAN | source code on GitHub

The package implements the methods described in these papers:

• Aki Vehtari, Andrew Gelman, and Jonah Gabry. (2017). Practical Bayesian model evaluation using leave-one-out cross-validation and WAIC. Statistics and Computing. 27(5), pp 1413-1432. published version | arXiv preprint

• Aki Vehtari, Daniel Simpson, Andrew Gelman, Yuling Yao, and Jonah Gabry. (2024). Pareto smoothed importance sampling. Journal of Machine Learning Research. 25(72):1-58 published version | arXiv preprint

rstantools

The rstantools R package provides tools for developing R packages interfacing with Stan.
website | CRAN | source code on GitHub

posterior

The posterior R package provides efficient conversion between many different useful formats of draws (samples) from posterior or prior distributions, consistent methods for operations commonly performed on draws, various summaries of draws in convenient formats, and lightweight implementations of state of the art posterior inference diagnostics.
website | CRAN | source code on GitHub

shinystan

The shinystan R package provides a graphical user interface providing interactive visual and numerical summaries of model parameters and convergence diagnostics for Bayesian models estimated using MCMC.
website | CRAN | source code on GitHub

Papers

Here’s an incomplete list of publications and preprints. You can find additional publications and other details on Google Scholar profile.

Published

• [2024] Pareto Smoothed Importance Sampling. Journal of Machine Learning Research (Aki Vehtari, Daniel Simpson, Andrew Gelman, Yuling Yao, Jonah Gabry)
• [2022] Fast methods for posterior inference of two-group normal-normal models. Bayesian Analysis (Philip Greengard, Jeremy Hoskins, Charles C. Margossian, Jonah Gabry, Andrew Gelman, Aki Vehtari)
• [2020] Efficient leave-one-out cross-validation for Bayesian non-factorized normal and Student-t models. Computational Statistics (Paul Bürkner, Jonah Gabry, Aki Vehtari)
• [2020] Approximate leave-future-out cross-validation for Bayesian time series models. Journal of Statistical Computation and Simulation (Paul Bürkner, Jonah Gabry, Aki Vehtari)
• [2020] Bayesian hierarchical weighting adjustment and survey inference. Survey Methodology (Yajuan Si, Rob Trangucci, Jonah Gabry, Andrew Gelman)
• [2019] Visualization in Bayesian workflow (with discussion). Journal of the Royal Statistical Society A. (Jonah Gabry, Daniel Simpson, Aki Vehtari, Michael Betancourt, Andrew Gelman)
• [2018] R-squared for Bayesian regression models. The American Statistician. (Andrew Gelman, Ben Goodrich, Jonah Gabry, Aki Vehtari)
• [2018] User-friendly Bayesian regression modeling. The Quantitative Methods for Psychology. (Chelsea Muth, Zita Oravecz, Jonah Gabry)
• [2018] Bayesian hierarchical models for predicting individual performance in soccer. Journal of Quantitative Analysis in Sports. (Leonardo Egidi and Jonah Gabry)
• [2017] Practical Bayesian model evaluation using leave-one-out cross-validation and WAIC. Statistics and Computing. (Aki Vehtari, Andrew Gelman, Jonah Gabry)
• [2017] Antiemetic prophylaxis as a marker of health care disparities in the National Anesthesia Clinical Outcomes Registry. Anesthesia & Analgesia. (Michael Andreae, Jonah Gabry, Ben Goodrich, Robert White, Charles Hall)
• [2017] A pragmatic trial to improve adherence with scheduled appointments in an inner-city pain clinic by human phone calls in the patient’s preferred language. Journal of Clinical Anesthesia. (Michael Andreae, Singh Nair, Jonah Gabry, Ben Goodrich, Charles Hall, Naum Shaparin)

Preprints

• Preoperative Exposure to Fine Particulate Matter and Risk of Postoperative Complications: A Single Center Observational Cohort Bayesian Analysis (John F Pearson, Cameron Jacobson, Calvin Riss, Matthew Strickland, Longyin Lee, Neng Wan, Tabitha M. Benney, Nathan L Pace, Ben Goodrich, Jonah Gabry, Cade Kartchner, Michael H Andreae)
• Multilevel Regression and Poststratification Interface: Application to Track Community-level COVID-19 Viral Transmission (Yajuan Si, Toan Tran, Jonah Gabry, Mitzi Morris, Andrew Gelman)
• Bayesian workflow (Andrew Gelman, Aki Vehtari, Daniel Simpson, Charles C. Margossian, Bob Carpenter, Yuling Yao, Lauren Kennedy, Jonah Gabry, Paul-Christian Bürkner, Martin Modrák)
• Using multilevel regression and poststratification to estimate dynamic public opinion. (Andrew Gelman, Jeffrey Lax, Justin Phillips, Jonah Gabry, Robert Trangucci)

Tutorials and educational materials

A selection of tutorial vignettes on practical topics in Bayesian data analysis.

Visualization

• Visual MCMC diagnostics
• Plotting MCMC draws
• Graphical posterior predictive checks

Bayesian regression modeling with rstanarm

• GLMs for Binary/Binomial Data
• GLMs for Continuous Data
• GLMs for Count Data
• Generalized (Non-)Linear Models with Group-Specific Terms
• Bayesian ANOVA Models
• Regularized Linear Models
• Ordinal Regression Models
• Hierarchical Partial Pooling for Repeated Binary Trials
• Modeling Rates/Proportions using Beta Regression
• Multilevel regression and poststratification

Predictive performance and model comparison

• Using the loo package (version >= 2.0.0)
• Bayesian Stacking and Pseudo-BMA weights
• Writing Stan programs for use with the loo package
• Leave-one-out cross-validation for non-factorizable models
• Approximate leave-future-out cross-validation for time series models
• Using Leave-one-out cross-validation for large data

Developing R packages with pre-compiled Stan programs

• Step by step guide for creating a package that depends on RStan