TLC: Privacy by Design AB Testing with Matt Gershoff

Hi TLCers!

It's that time again. Are you ready to be #mattgershoffed? Join me in welcoming back TLC favorite, Matt Gershoff, co-founder of Conductrics, to the TLC stage!

Y'all have been asking for some deep technical stuff. More math! More complex concepts! More advanced concepts! Challenge us! Well, who better to teach us than Matt? And - by the time we're all done, you know as well as I do - your confusion will turn into clarity. And that's what it means to be #mattgershoffed. As much as he hates that hashtag, those of us who know him, love it. Because we've been learning from him for years.

Topic:
Privacy by design is encoded in both the GDPR/CPRA legislation. This talk will quickly review the 7 privacy by design principles. We will then show how data thoughtfully aggregated can be stored in equivalence classes which: 1) naturally leads to use of k-anonymity as a simple measure of privacy; and 2) enables extremely efficient computation of many of the statistical methods used in AB Testing (t-tests, f-tests, ANOVA, ANCOVA). After this talk you should have a general understanding of privacy by default, K-anonymization, and the equivalence of OLS regression results based on microdata (data at individual level) and aggregated data (data at the experiment level).

Join the conversation Oct 11.

9 AM Pacific | 10 AM Mountain | 11 AM Central | 12 PM Eastern | 5 PM London