These are links to resources related to questions asked during the workshop. Let us know if we forgot anything!
- "Bayesian Data Analysis 3rd", Andrew Gelman et al., book, webpage
- "Statistical Rethinking", Richard McElreath, webpage, youtube
- Good first case study: Hierarchical partial pooling for repeated binary trials
- Past Stancon tutorials
- The golf example Andrew presented
- "Visualization in Bayesian workflow", Jonah Gabry et al., paper code
- "The prior can generally only be understood in the context of the likelihood", Andrew Gelman et al., link
- "Diagnosing Biased Inference with Divergences", Michael Betancourt, link
- "QR Reparameterization", Stan Manual, link
- "Fitting The Cauchy Distribution", Michael Betancourt, link
- "Reparameterization", Stan Manual, link
- "Hamiltonian Monte Carlo for Hierarchical Models", Michael Betancourt et al., link
- "Finite Mixtures", Stan Manual, link
- "Identifying Bayesian Mixture Models", Michael Betancourt, link
- "Multivariate priors for hierarchical regressions", Stan Manual, link
- "Gaussian processes", Stan manual, link
- Relationship to causal inference
- Missing data techniques in the Stan manual
- More examples of integrating out parameters
- "Zero-Inflated and Hurdle Models", Stan manual, link
- Explicit support for univariate truncated distributions, manual chapter
- Truncated multivariate normal, Stan forum post, link to paper+code
- Single computer parallelization via
reduce_sum
(easier), tutorial, user's guide, functions reference - Multiple computer parallelization via
map_rect
(harder), tutorial, user's guide, functions reference
- "MRP in RStanarm", Lauren Kennedy, link
- Deep interactions in MRP Ghitza, Gelman paper
- Structured priors for MRP: Gao, Kennedy, Simpson, Gelman paper
- Case study: A Bayesian Approach to Sequential A/B Testing: Multi-Armed Contextual Bandits in Stan" by Bob Carpenter (code)[https://github.com/stan-dev/example-models/tree/master/knitr/bandits)
- "Model-based Inference for Causal Effects in Completely Randomized Experments", Joon-Ho Lee et al. link
- "Adapting Structural Equation Modeling Fit indices into Bayesian SEM, Psychological Methods", Mauricio Garnier-Villarreal et al., discourse paper
- "Bayesian hierarchical spatial models: Implementing the Besag York Mollié model in stan", Mitzi Morris et al., link
- Multiple Output Gaussian Processes, Rob Trangucci, link
- What does
y ~ normal(mu, sigma);
mean?, "Sampling Statements", Stan Manual, link - How do constraints work
real<lower = 0.0> sigma;
?, "Constraint Transforms", Stan Manual, link
- "Matrices, Vectors, and Arrays", Stan manual, link
- Matrix multiplication vs. elementwise multiplication, Stan manual, link
- "Reproducibility", Stan manual, link
- Approximate leave-one-out cross validation in Stan: loo package vignette
- Writing Stan programs for use with the loo package: loo package vignette
- Aki’s model selection webpage: https://avehtari.github.io/modelselection/
- Aki’s Cross Validation FAQ https://avehtari.github.io/modelselection/CV-FAQ.html
- "Using Stacking to Average Bayesian Predictive Distributions (with Discussion)", Yuling Yao et al., link
- "Bayesian Stacking and Pseudo-BMA weights using the loo package", Aki Vehtari and Jonah Gabry, link
- Andrew's blog post: The folk theorem of statistical computing
- Folk theorem - something people know is true but can’t explain
- Figure 7 in "The No-U-Turn Sampler: Adaptively Setting Path Lengths in Hamiltonian Monte Carlo", Matt Hoffman, link
- Table 2 in "Faster estimation of Bayesian models in ecology using Hamiltonian Monte Carlo", Cole C. Monnahan et al., link
- Gibbs vs. Stan for IRT modeling, "Fitting Bayesian item response models in Stata and Stan", link
- These largely replace traceplots and autocorrelation plots)
- "New R-hat and ESS", Aki Vehtari, link
- "Rank-normalization, folding, and localization: An improved Rˆ for assessing convergence of MCMC", Aki Vehtari et al., link
- "Rank-normalization, folding, and localization: An improved Rˆ for assessing convergence of MCMC (online appendix)", Aki Vehtari et al., link
- "MCMC using Hamiltonian dynamics", Radford Neal, link
- "A Conceptual Introduction to Hamiltonian Monte Carlo", Michael Betancourt, link
- "Faster estimation of Bayesian models in ecology using Hamiltonian Monte Carlo", Cole C. Monnahan et al., link
- Monte Carlo algorithm animations, Chi Feng, link
- Page 279 of "Bayesian Data Analysis", Andrew Gelman et al., link
- "The Fundamental Incompatibility of Hamiltonian Monte Carlo and Data Subsampling", Michael Betancourt, link
- The variational inference algorithm in Stan is still in development. There are various papers and commentaries on it
*Yes But Did It Work? Evaluating Variational Inference: https://arxiv.org/pdf/1802.02538.pdf
- Jonathan Huggins’s work on Variational Bayes:
- Recent discussion: forum post