ISSUE 302 ·   September 8, 2020        

Insight

Fixing the analytics models that COVID-19 broke

COVID-19 has upended life as we knew it, and our new behaviors are wreaking havoc on some analytics models that rely on historical data. How can leaders enable teams to get analytics back on track?
McKinsey & Company

Information, incentives, & goals in election forecasts

This new paper explores the challenges in evaluating, understanding, and communicating election predictions and why the highly regarded forecasts from FiveThirtyEight and The Economist aren't the same.
Andrew Gelman, Jessica Hullman, Christopher Wlezien

The True Colors of America’s Political Spectrum

They may not look like the lines on a map but political boundaries are hardly lost when the world is viewed from a satellite. Brilliant.
New York Times

Sponsored Link

Future-Proof Your Career, While You Work

The Global Master of Management Analytics from Smith School of Business at Queen’s University is a 12-month program that can be taken from anywhere in the world.  Master the essential strategies for applying analytics to business needs in this ground-breaking program.

Tutorials, Projects & Opinions

Mathematical Modeling of Football ⚽️

This comprehensive course shows how to work with ⚽️ data and understand the game using math and stats.  It's taught by David Sumpter, author of Soccermatics, with guest lectures from leading practitioners and researchers around the world. Registration is closed but all the videos, lecture notes and exercises are available for free. 
Uppsala Universitet

Column Names as Contracts

Great post by Emily Riederer that shows how to simplify data wrangling and management using controlled vocabularies. This is a low-tech approach that's well thought-out and easy to implement. 
Emily Riederer

Traffic prediction with Graph Neural Networks

Nice introduction to a practical problem solved with Graph Neural Networks. It's pretty high-level and describes a technique that improves the accuracy of Google Map ETAs by up to 50%.
DeepMind

Ray Summit: Scalable Python & ML for everyone

See how companies like AWS, Microsoft & JP Morgan use Ray's distributed framework and ML libraries to build AI applications at any scale. Join Ray Summit (Sep 30-Oct 1) and learn from David Patterson, Michael Jordan, Oriol Vinyals, Ion Stoica & more. Register now to join the livestream or watch the sessions on-demand — it's free.
// sponsored

Code & Tools

•  fastcore - Python goodies to make your coding faster, easier, and more maintainable

•  jamovi - open statistical software to bridge the gap between researcher and statistician

•  Matplot++ - data visualization library for C++

•  arangodb - a scalable open-source multi-model database natively supporting graph, document and search.

Resources

Kickstarter SQL Style Guide

Great guide for writing clean, clear, and consistent SQL. Even if you don't agree with all the details, this is a worthwhile starting point for discussion and for writing a SQL Style Guide for your own team. 
 GitHub | Fred Benenson 

Visualizing Uncertainty - A Short Introduction

For a lot of reasons, uncertainty is one of the hardest things to present clearly in a chart or visualization. This 60 page booklet aims to change that. Covers current research, challenges and methods.
Analysis under Uncertainty for Decision-Makers Network (AU4DM)

Data Elixir is curated and maintained by Lon Riesberg. If you need help on a data project or have a suggestion for the newsletter, reply back to this email or grab a spot on my calendar >>