Subject: New book: Hamiltonian Monte Carlo in Stata (bayeshmc package)
Dear Statalist:
I am writing to let the list know about the publication of a new book
documenting the bayeshmc Stata package:
Ben A. Dwamena (2026). Hamiltonian Monte Carlo in Stata:
Bayesian Regression Modeling with the bayeshmc Package.
BennyBeauBooks, East Lansing, Michigan. Paperback, 380 pages.
ISBN 979-8-9951337-0-4. USD 59.95.
https://a.co/d/0b2gWygH
The bayeshmc package brings gradient-based Hamiltonian Monte Carlo (HMC)
and the No-U-Turn Sampler (NUTS) to Stata through the Stan probabilistic-
programming language via CmdStan. The estimation syntax follows Stata
conventions; the underlying Stan program is generated automatically. The
book documents the package and, more broadly, offers a practical
reference for Bayesian regression modeling in Stata.
Coverage:
- Bayesian foundations and MCMC; principles of Hamiltonian dynamics and
NUTS
- Prior specification, convergence diagnostics (R-hat, ESS, divergences,
tree depth), and posterior visualization
- More than forty regression chapters, including linear, generalized
linear, ordinal, multinomial, censored and truncated, panel-data, and
multilevel mixed-effects models
- Covariance priors for random effects; model comparison via WAIC and
PSIS-LOO
- A chapter on Bayesian diagnostic test accuracy meta-analysis
- Side-by-side comparison with Stata's built-in bayes: prefix
throughout, indicating when each is preferable
Each chapter presents the model specification, prior recommendations, a
worked example with data and code, and the auto-generated Stan program.
Appendices cover installation, diagnostics, and troubleshooting.
Hardcover and Kindle editions will follow in the coming weeks.
Comments and corrections are welcome.
Best regards,
Ben A. Dwamena, MD
Clinical Associate Professor Emeritus of Radiology
Division of Nuclear Medicine and Molecular Imaging
University of Michigan, Ann Arbor
[email protected]
Dear Statalist:
I am writing to let the list know about the publication of a new book
documenting the bayeshmc Stata package:
Ben A. Dwamena (2026). Hamiltonian Monte Carlo in Stata:
Bayesian Regression Modeling with the bayeshmc Package.
BennyBeauBooks, East Lansing, Michigan. Paperback, 380 pages.
ISBN 979-8-9951337-0-4. USD 59.95.
https://a.co/d/0b2gWygH
The bayeshmc package brings gradient-based Hamiltonian Monte Carlo (HMC)
and the No-U-Turn Sampler (NUTS) to Stata through the Stan probabilistic-
programming language via CmdStan. The estimation syntax follows Stata
conventions; the underlying Stan program is generated automatically. The
book documents the package and, more broadly, offers a practical
reference for Bayesian regression modeling in Stata.
Coverage:
- Bayesian foundations and MCMC; principles of Hamiltonian dynamics and
NUTS
- Prior specification, convergence diagnostics (R-hat, ESS, divergences,
tree depth), and posterior visualization
- More than forty regression chapters, including linear, generalized
linear, ordinal, multinomial, censored and truncated, panel-data, and
multilevel mixed-effects models
- Covariance priors for random effects; model comparison via WAIC and
PSIS-LOO
- A chapter on Bayesian diagnostic test accuracy meta-analysis
- Side-by-side comparison with Stata's built-in bayes: prefix
throughout, indicating when each is preferable
Each chapter presents the model specification, prior recommendations, a
worked example with data and code, and the auto-generated Stan program.
Appendices cover installation, diagnostics, and troubleshooting.
Hardcover and Kindle editions will follow in the coming weeks.
Comments and corrections are welcome.
Best regards,
Ben A. Dwamena, MD
Clinical Associate Professor Emeritus of Radiology
Division of Nuclear Medicine and Molecular Imaging
University of Michigan, Ann Arbor
[email protected]
