Why not t-test p99 latency? Describe a valid alternative.
This tests your grasp of statistical test assumptions. A good answer explains why p99 violates t-test normality, then outlines a resampling method like bootstrapping to build a confidence interval on the *difference* of p99s.
This tests your understanding of statistical test assumptions, specifically the Central Limit Theorem's limits. A great answer first explains why the distribution of a sample percentile (like p99) is not Gaussian, making t-tests invalid. Then, it outlines a resampling method like bootstrapping: repeatedly sample with replacement from both groups, calculate the p99 difference each time, and use the resulting distribution to find a confidence interval.
Read the original → allendowney.com
- #a/b testing
- #statistics
- #latency
- #metrics
- #p99
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.