In this week’s brief, a continuation of our “Statistical Thinking” series, we reflect on three “myths” in data science and statistics, and spotlight our ten-course Social Science Statistics certificate program. You can get started with either of these courses:
- Aug 7- Sep 4: Survey Design and Sampling Procedures
- Oct 2 – 30: Regression Analysis
See you in class!
Peter Bruce
Three Myths in Data Science
Small Ball*
In this mature age of digital marketing, companies have developed finely honed engines of automated and target promotion that factor in individual preferences and behavior. The idea is to add small increments to revenue and profit. The system evolved in a stable era of economic expansion, but the automated promotions are now running afoul of the reality of a world disrupted by Covid. Online wine merchants use your purchase history and promotional inducements to encourage you to buy, but, when you do click on the “buy” button, they can’t tell you when they will deliver, due to shipping and distribution disruptions. This is a problem when a resident must be at home to sign for delivery. Airlines upgrade their best customers (and deserving idle crew) gratis to first class, which results in first class cabins being full, while economy class sits mostly empty. With free drinks and snacks in first class gone as part of the Covid regime, a crowded first-class cabin suddenly becomes less attractive. “Downgrading” to an empty section of coach was not something the algorithm was used to. Even if they succeeded in moving out of coach, the favored customer would end up crowded in with other favored customers in the front “extra-leg-room” rows of coach. The largest U.S. airline has ended up overriding its finely tuned seating algorithms that had evolved over years with fixed rules about cabin density. It will be interesting to see if they can find a way to monetize solitude.
*The term “small ball” refers to a style of baseball play in which managers don’t wait for low probability, high-impact events (3-run home-runs, for example), but rather cobble together strings of lower-impact but higher probability scoring efforts (e.g. stolen bases, sacrifices to advance runners). We use it here to refer to the use of machine learning algorithms to incrementally improve revenue, profit, etc.
Certificate Spotlight
Our ten-course Social Science Statistics certificate program gives you lots of flexibility in both topical coverage and scheduling. It is your chance to take a deep dive into statistical modeling methods that help you better understand the relationships in your data. Enrollees in this certificate program include professionals working at
- Government statistical agencies
- Higher ed institutions
- Consumer goods marketing
- Local and regional government agencies
Compare the in-depth coverage in this 10-course program to any graduate program, and then compare the cost – just $5.000 for the whole program.
You can get started with either of these courses:
- Aug 7- Sep 4: Survey Design and Sampling Procedures
- Oct 2-30: Regression Analysis
See you in class!
