R-bloggers

Ah great, the new book on Bayesian workflow by Andrew Gelman, Aki Vehtari, Richard McElreath I knew they were working on is about to appear!  With entries from several coauthors and half of the chapters on case studies. I have not (yet) looked at its c...

A chance comment from a colleague led to today’s post. Competition for research grants in the UK has become more intense recently. Success rates have dropped, and this has been attributed to more applications, against a backdrop of a flat investment in science by the government. My colleague thought ...

maestro has officially graduated to stable release with version 1.0.0 back in January 2026 and now its latest version 1.1.0. This marks a commitment to maintaining a stable API and increased reliance on using maestro in production. In our e...

This wonderful post from Jonathan Caroll give me idea to play with ggplot2 code and scico color palettes: Also see this toot from Stefan Siegert, inspiring! I took the code from the toot mentioned above, modified it, and used it to have a littl...

Translating things between languages reveals how each language approaches different design trade-offs, and I believe it’s a useful exercise. Having something to translate is the first step. I found a plot I wanted to generate, and some code that reproduced it, so off we go! I don’t recall ...

R 4.6.0 (“Because it was There”) is set for release on April 24th 2026. Here we summarise some of the more interesting changes that have been introduced. In previous blog posts, we have discussed the new features introduced in R 4.5.0 and earlier versions (see the links at the end of this post). ...

I wanted to know: After a little bit of searching, I couldn’t find any answers. So I decided to use R to retrieve the necessary info from Uniprot and calculate it myself. I thought I’d post it here in case it’s useful for others. Human We’ll ...

Most health care economics models are constructed from the perspective of a managed health care system such as those offered in Canada and several European countries, or from the perspective of some other third party such as an insurance company...

1 A model can run and still be fundamentally wrong Many time series models fail before they even begin. Not because the software crashes. Not because the code is wrong. But because the data entering the model violate one of the most impor...

dplyr verbs are descriptive: let’s make them more verbose! Yet another pipe for R. Motivation In SAS, every DATA step prints a log: NOTE: There were 120000 observations read from WORK.SALES. NOTE: 7153 observations wer...

With field experiments, studying the correlation between the observed traits may not be an easy task. For example, we can consider a genotype experiment, laid out in randomised complete blocks, with 27 wheat genotypes and three replicates, where sev...

Working with LLMs in Practice Large Language Models are becoming part of everyday data science work. But using them through chat interfaces is only one part of the picture. In this upcoming webinar, we focus on how to work with LLMs programmatica...

It’s often the way. I posted recently about how to pace a marathon and very quickly received feedback that would’ve improved the original post. Oh well, no going back. This is take two. So, we have a dataset of all runners from the 2025 New York City Marathon. We ...

I’m often asked if using AI coding agents saves time. Yes they write code very quickly and can complete entire ecological data analyses. Do agents really help when the deadlines are approaching? Do agents really help when the deadlines are ...