Kernel smoothing for calibration estimation
I’ll show some early calibration plots of my calibration tracking, and show how I’ve used a few smoothing techniques to produce useful early charts.
About me
Hi, I'm Richard, a data scientist-y person. I like solving interesting, challenging problems in fun, pragmatic ways. I'm currently a Principal Data Scientist at Mesh-AI. Previously, I was the Head of Data & Analytics at what3words, a Data Scientist at Zulzi (predominantly building recommender models), and a consultant at Opti-Num Solutions.
I write here about some of the things that interest me.
Writing
"Cheating" at Codenames with vector embeddings
Have you ever wondered whether there’s an optimal strategy for the card game Codenames? This post details my quest to find a (near) optimal solution, which is tec...
A/B test analysis with bootstrapping
In this post I’ll give a hands-on walkthrough showing you how to analyse the impact of an A/B test using bootstrap sampling. I’ll also compare this to the traditi...
Intro to bootstrapping for confidence bounds
I’ll show you how to use bootstrapping, a powerful sampling technique which can be used to estimate confidence bounds, p-values and distributions for samples with...