How do I approach Statistics & Math interview questions?

Statistics & Math questions require understanding of core concepts and practice. PracHub provides solutions with explanations to help you master statistics & math interviews.

What difficulty level is this interview question?

This is a medium difficulty Statistics & Math question, commonly asked during Technical Screen rounds at Uber.

What role is this question designed for?

This question is commonly asked for Data Scientist candidates at Uber during technical interviews.

Should Uber double member discounts? | Uber Interview Question

Q: Should Uber double member discounts?

This question evaluates competency in causal inference, experimental design, statistical power and sample-size analysis, metric definition, and two-sided marketplace economics.

Uber is considering increasing the member discount on rides from 5 percent to 10 percent. This can affect rider demand, driver supply, marketplace balance, and overall unit economics in a two-sided marketplace.

Answer the following:

What are the main potential benefits and costs of this policy change?
Which primary success metrics, guardrail metrics, and diagnostic metrics would you track? Consider rider conversion, trips per member, gross bookings, take rate, contribution margin, wait time, cancellation, driver earnings, and non-member cannibalization.
How would you design an experiment to evaluate this change in a marketplace with interference?
If you propose a switchback design, what assumptions must hold for unbiased inference? In what real ride-hailing scenarios could those assumptions fail?
Which parameters determine sample size and minimum detectable effect? How does clustering or switchback randomization change the calculation?
How would you decide how long the experiment should run?
Suppose the experiment is planned for two months, but halfway through the p-value for the primary metric is 0.04. Should the company stop early and launch? Why or why not?

Answer the following:

What are the main potential benefits and costs of this policy change?
Which primary success metrics, guardrail metrics, and diagnostic metrics would you track? Consider rider conversion, trips per member, gross bookings, take rate, contribution margin, wait time, cancellation, driver earnings, and non-member cannibalization.
How would you design an experiment to evaluate this change in a marketplace with interference?
If you propose a switchback design, what assumptions must hold for unbiased inference? In what real ride-hailing scenarios could those assumptions fail?
Which parameters determine sample size and minimum detectable effect? How does clustering or switchback randomization change the calculation?
How would you decide how long the experiment should run?
Suppose the experiment is planned for two months, but halfway through the p-value for the primary metric is 0.04. Should the company stop early and launch? Why or why not?

Should Uber double member discounts?

Quick Overview

Solution

Submit Your Answer

Should Uber double member discounts?

Quick Overview

Solution

Submit Your Answer