R-bloggers

Ask a large language model a question about your data, and you’ll probably get a confident answer. Ask it again, and you might get a different one. Sometimes, neither answer is correct. That’s a problem if you’re building an AI ...

I recently learned about an interesting way of reducing the dimension of a data set. I’d like to illustrate it with a data set of about 550 historical battles in the UK, extracted from Wikidata using a SPARQL query. No doubt some readers will have seen...

Summary Who this is for: Data scientists and analysts who need to estimate what an intervention does (a call, a discount, a campaign) and want to know when causal ML methods are worth using over LM, GLM, and GBM. The basis is nine simulations w...

It’s really easy to make your own agent. You should do it once, just to understand how this technology works. LLM agents are becoming pervasive, with Meta’s Muse perhaps the first consumer grade agent. They are only going to get more common. If ...

Exploring the integration of caretS3 and ahead for tabular and time series machine learning tasks.

This blog post is about me trying to make a better version of an existing diagram that I think is important, but difficult to read in the version it keeps popping up on my Bluesky feed. Robin J Brooks of the Brookings Institution has been using a usef...

What would happen if we peeled the Earth like an orange? The interrupted Goode homolosine projection gives us something close to that! This equal-area projection preserves relative areas while splitting the world map into lobes. Add a few shades of orange and our world is ready to peel.

Zhenguo Zhang's Blog https://fortune9.netlify.app/2026/10/23/python-simpsons-paradox/ - Introduction Simpson’s paradox occurs when the direction of an association reverses after data from several groups are combined. For example, A can have a higher success rate than B in every subgroup, yet a lower success rate overall. In this ...

Following this previous post and this discussion on Mastodon, I wrote a function that wraps a combination of functions from the ‘terra’ package to project a SpatRaster or SpatVector layer while automatically centering the projection on that layer, to minimize … Continue reading →

When forecasting high-volatility financial assets like Bitcoin (BTC-USD), standard point-prediction models—whether classical ARIMA or deep neural networks—often fall short. A point estimate tells you where the model expects the price to go, but it completely ignores market regime shifts and changing risk levels. In this article, we construct ...

1 Documentation was always the point TKCat (Tailored Knowledge Catalog) is built on a simple idea: a knowledge resource is more than a set of tables. In TKCat, each resource is a modeled database (MDB) that contains: a general description of t...

REX d’une firme minimale : auditer sa veille IA, protéger ses notes et mesurer les limites du contrôle humain.

I often need to get mainland from country / administrative region polygons, so I finally decided to write a function to help automate it. It selects the N largest feature(s) in a polygons or lines map. Here’s the code: The … Continue reading →