How does CUPED increase the statistical power of an experiment?
Tests your grasp of variance reduction. Explain CUPED as ANCOVA, using pre-experiment data (X) to remove predictable noise from the outcome (Y). Effectiveness depends on correlation (rho), reducing variance by (1-rho^2).
Tests your grasp of variance reduction in A/B testing. A great answer defines CUPED as ANCOVA, using highly correlated pre-experiment data (X) to reduce variance in the outcome metric (Y). It constructs an adjusted metric by removing the predictable noise from Y, achieving a variance reduction of (1-rho^2). A red flag is describing a simple difference-in-difference (Y-X) approach, which can be less effective or even harmful.
Read the original → zetyra.com
- #ab testing
- #statistics
- #metrics
- #data science
Get five bites like this every day.
Tezvyn delivers a daily feed of 60-second tech bites with quizzes to lock in what you learn.