ii. Explain clearly why Uber pursued a switchback experiment
rather than an A/B test to evaluate the impact of a 5-minute wait
as compared to a 2-minute wait on various performance metrics in
Boston?