Randomized and quasi-randomized 'neural' networks for supervised learning

The use of high-frequency data has gained widespread attention and popularity in contemporary financial research thanks to its potential to capture minute fluctuations and provide detailed insights into the behavior of financial markets. However, analy...

Self-Guided Learning: The Ever Evolving Wordle Guess Generator The best way to learn a programming language is to code. Obvious, but sometimes it can be daunting to start a project when you are a novice. Influenced by my time at a start-up, I’ve ...

This is the current version of the word game guesser. I discussed how I used this project to hone my coding skills in the companion blog post to this one. I’m not going to walk through this in much detail, but I’m going to point out some of ...

Creating custom user interfaces with {shiny} and {bslib} has never been easier. Using {bslib} it is also incredibly simple to choose and switch between versions of the underlying Bootstrap library that powers the UI, to use many of the Bootswatch theme...

We are happy to announce our fifth series of workshops, to take place in 2023. Between May and June, we are proposing again two learning paths: R & Shiny, with few variations compared to last year. Like always, the workshops composing a path can...

In a previous blog post I discussed how we can compute rank dependent utility in R. Though this function was quite exciting for myself as computing rank dependent utilty by hand is quite labor intensive, I saw the function as limiting in the sense that I was suppressing all the ...

Together with Andreas Küpfer I spent the last months writing and thinking about how to best explain BERT (using huggingface), apply it to a typical social science problem, and show the functionality of explainable AI. When working with large language m...

I’ve recently been playing around with classification models, specifically on data sets with a skewed class distribution. In imbalanced classification problems, one class occurs infrequently. The minority class is often the class of interest (think fraudulent transaction, positive disease diagnosis, or intruder detection). Sometimes these applications are framed as ...

Here’s a short post on why one should consider developing a segmented model when building credit risk scorecards to help optimize approval rates. While segmentation offers a variety of benefits, this post aims to offer a different perspective. I ...

I won’t wax long and poetic here since I’ve already posted the experiment that has all the details. TL;DR: there are still only ~90-ish 📦 in the WebR WASM “CRAN”, but more are absolutely on the way, including the capability to build your own CRAN and dev packages ...