📌 Pinned posts
- None of the above! – This is a special post existing to allow users to post comments and questions about other meta-analytic models, software, and evidence base problems that are not dealt with elsewhere on this website. As time passes, if some of the topics become more popular, we will move the comments to a new post and write about […]
- Software comparison table – When you are deciding on the software you would prefer to use for Bayesian meta-analysis, there are several aspects to bear in mind. We thought it would be useful to put them into one table. This appears at the back of the book as an appendix, but we can extend and update it here. Software […]
- Events, courses, conferences, etc – This post is the place where you can add information about events. Just add a comment below and put in the details. Remember to give dates, times with time zone, and a link to more information. Please don’t make your comment too long; readers can click your link to find out more.
- How to install your Bayesian software – The first problem in Bayesian analysis is generally how to install the software. We didn’t want to give advice on this in the book itself because requirements change from time to time. Here, we provide links to official advice so that you can hear it directly from the developers. The book includes BUGS, JAGS, Stan, […]
All other posts
- The snow leopards example – Code accompanying Chapter 2 of our book “Bayesian Meta-Analysis: a practical introduction”. This is a simple example with an imaginary data set. It allows the reader to explore ideas of models, likelihood and posterior probability.
- Simulating correlated data for time points – Learn how to simulate data from studies with multiple time points and other correlation structures. Make better informed decisions about meta-analysis and related methods by realistic simulation.
- Simulate randomised controlled trials to learn about meta-analysis – Simulate data from randomised controlled trials in order to learn about statistical methods for meta-analysis.
- Using Stata 19 evaluators for Bayesian meta-analysis – This is a summary of a workshop Robert gave at the Italian Stata Users’ Group in Milan in September 2025. Stata version 19, released in 2025, introduced new additional flexibility to its Bayesian capabilities which makes Stata now fully flexible for all the BMA models that we discuss in the book. This is a really […]
- Going beyond the normal assumptions in random effects models in meta-analysis? – Most meta-analyses are based on the random-effects model, the standard go-to framework based on the assumption that each study estimates a different “true” effect and that these true effects follow a normal distribution. This assumption underpins nearly every random-effects analysis performed in clinical and public-health research. These questions motivated Panagiotopoulou et al. (Panagiotopoulou, K., Evrenoglou, […]
- Online Bayesian MA training with the RSS – If you would prefer to get started in Bayesian meta-analysis with a two-day online course on the topic, then consider the course that Robert has been running with the Royal Statistical Society since 2020.
- Using the bootstrap to estimate Monte Carlo error of posterior quantiles – Learn how to estimate the Monte Carlo standard error for any quantile of a Bayesian posterior sample. This shows the margin of error that might undermine decision making.
- Simulate data to learn about inference, Using R – This post provides R code to investigate, hands-on, ideas of statistical inference, to prime you for moving on to thinking about models in terms of likelihoods. It expands pp7-8 of our book.
- All binary pairwise models in cmdstanr – Here is R code you can use to learn about the basic pairwise models for binary outcomes, with simple simulated data. There are combinations of {common effect, random effects} x {contrast-based, arm-based} with examples of heuristic t likelihood and alternative specification of arm likelihoods. On the end, we show normal equivalents with reasonably large n […]
- When study statistics might have a typo – This post expands on subject matter of Chapter 5 of our book, on extracting study statistics for meta-analysis. Uncertainty about a published statistic is one of the most common reasons to consider going Bayesian in your meta-analysis. Because probability can represent any uncertainty, we can use a prior distribution for the uncertain statistic instead of […]