John Elder will deliver the virtual talk “Top 3 Things I’ve Learned in 3 Decades Of Data Science” to the students in the Department of Statistical Science at Duke University on February 17, 2021.
The three most important analytic innovations I’ve seen in three decades of extracting useful information from data have to do with: Ensemble models, Target Shuffling, and Cognitive Biases. Ensembles are sets of competing models that often combine to be more accurate than the best of their components. Target Shuffling is a resampling method that corrects for “p-hacking” or the “vast search effect” where spurious correlations are frequently seized upon by modern methods’ ability to try millions of hypotheses. And the third, Cognitive Biases, is the dawning understanding of how deeply flawed our own reasoning naturally is. This especially reveals how doomed our projects will be unless we seek out — and attend to – external critique.