Next Newsletter

Find the quietest routes in your state and get ready to hit the open road.

Annotated talk slides

*Most words are in English.

AI terms explained simply.

A Comprehensive Overview of Prompt Engineering

Part 8 of the GLM and causal inference series.

Part 9 of the GLM and causal inference series.

Watch 24 hours of an American day, and the invisible crisis hiding in plain sight

Loading links to Bluesky, Facebook, Instagram, and Substack takes far longer than to other sites. We built a tool so you can check any domain you like

Where does the Capital Spending Database go from here?

Previous on this blog: 10 years of rio, rio 1.0.0

For posit\:\:conf(2023) I spent a ridiculous amount of time formatting my quarto slides and figuring out how to host them (easily) on github pages. This post summarises that process.

I visit Emil Hvitfeldt’s excellent resource cataloguing ggplot2 colour palettes at least once a week, and I usually go back to the same two packages (Nord and LaCroixColoR). The last few times I visited I’ve noticed a link to Simon J’s tutorial for making your own package.

I really can’t believe I spoke at posit::conf(2023)!. Virtual attendance posit::conf(2023) used discord and streamed talks to provide a virtual stream and I have to say I think it worked really well.

I do not remember when or how I first encountered the CSS Scroll Snap technique, but when I saw it, one thing came to my mind immediately: can we use this technique to make slides? (Spoiler: the …

The analysis of spatial and spatio-temporal point patterns is becoming increasingly necessary, given the rapid emergence of geographically and temporally indexed data in a wide range of fields. Non-parametric point pattern methods are a highly adaptable approach to answering questions about the real-world using complex data in the form of collections of points. Several methodological advances have been introduced in the last few years. This paper examines the current methodology, including the most recent developments in estimation and computation, and shows how various R packages can be combined to run a set of non-parametric point pattern analyses in a guided and intuitive way. An example of non-specific gastrointestinal disease reports in Hampshire, UK, from 2001 to 2003 is used to illustrate the methods, procedures and interpretations.

About a thing that makes finding some information really hard

Discover undocumented features of GitHub: open repos in online VSCode, add experimental features, regex code search, and more!

For their research on tattoos and choice, Bradley Ruffle and Anne Wilson provide the dataset for perusal: Survey and experimental evidence documents discrimination against tattooed individuals in t…

“Anscombe’s quartet” and the “datasaurus dozen” demonstrate the importance of visualizing data

I have a confession to make: I am now a Python user. Don't judge me, join me! In this post, I introduce Python for data analysis from the perspective of an R (tidyverse) user.

Let LLMs run code locally. Chat from your terminal

Using modern data tools to discover modern opportunities for exploration

Creating a chatbot that leverages the power of Langchain, OpenAI's ChatGPT, Pinecone, and Streamlit Chat.

The shift beyond China threatens to push up prices for consumers.