About the book
Synopsis
For the Stata analyst whose model has outgrown the canned families, The Stan Workbench closes the gap between Stata and Stan. The long-form companion to the stanrun suite, it lets you write a model in Stan, fit it with Hamiltonian Monte Carlo, and carry every intermediate result — draws, diagnostics, predictive checks, and information criteria — back into Stata, where your data, do-files, and reporting already live.
A modular, transparent pipeline of small commands takes an analysis from a Stan program through sampling to diagnostics, predictive checks, and model comparison. Developed across thirty-five chapters in seven parts with four reference appendices, it is a bench manual for the researcher who needs full control: Gaussian processes, item-response and joint longitudinal-survival models, hidden Markov models, mixtures, and custom likelihoods that the prefix commands cannot express.
← Back to all books
BennyBeauBooks
MIDAS: Meta-Analytical Integration of Diagnostic Accuracy Studies
Hamiltonian Monte Carlo in Stata