A bunch of readings and a new book on Bayesian meta-analysis

Robert Grant writes:

My Bayesian Meta-Analysis book is now available. There is a discount code “BMA25” for 25% off until 6 September.

The website also has code (of all software options), examples and discussion.

There’s a lot yet to be settled in this topic, which we only just touch on briefly in a closing chapter, for example with Bayesian MAs of Bayesian studies, or RCTs borrowing external control information, or Gaussian processes, etc etc. Even small-m and small-p tasks in this space can be highly correlated and benefit from Stan.

I haven’t seen this new book, but the topic is important!

Here are some things my collaborators and I have written on Bayesian meta-analysis:

– Section 5.6 of Bayesian Data Analysis

Meta-analysis with a single study

Water Treatment and Child Mortality: A Meta-analysis and Cost-effectiveness Analysis

Priors for hyperparameters in meta-analysis

Exploring some questions about meta-analysis (using ivermectin as an example), with R and Stan code

The real problem of that nudge meta-analysis is not that it includes 12 papers by noted fraudsters; it’s the GIGO of it all

No, I don’t like talk of false positive false negative etc but it can still be useful to warn people about systematic biases in meta-analysis

The p-curve, p-uniform, and Hedges (1984) methods for meta-analysis under selection bias: An exchange with Blake McShane, Uri Simonsohn, and Marcel van Assen

Constructing an informative prior using meta-analysis

Meta-analysis, game theory, and incentives to do replicable research

Using Bayesian meta-analysis to adjust for bias in experiments and observational studies

All these free readings, along with that 25%-off book, should give keep you busy for awhile!

2 thoughts on “A bunch of readings and a new book on Bayesian meta-analysis

  1. Thanks for posting! We are populating the website/blog over the summer so that all the models in the book will be there in all the software options (the ones that are compatible anyway). The software options we talk about in detail in the book are: Stan (obviously), brms, BUGS/JAGS, R package bayesmeta, Stata, and JASP. We also want to expand into PyMC, then R packages RBesT and multiNMA in due course. Essentially, we are opinionated but we describe all the reasonable options in use, e.g. we don’t like vague/default priors and we don’t find use for Bayes factors, but we describe them anyway.

Leave a Reply

Your email address will not be published. Required fields are marked *