Amazon Statistics & Math Interview Questions

Amazon Statistics & Math interview questions focus on applied inference, experiment design, probability, and mathematical reasoning used to drive product decisions. Unlike purely theoretical exams, Amazon's screens tend to evaluate your ability to translate statistical results into business recommendations: designing and interpreting A/B tests, selecting appropriate estimators, reasoning about bias and variance, and quantifying uncertainty for stakeholders. Expect questions rooted in realistic datasets and product trade-offs, often combined with SQL or simple coding to show you can manipulate data and validate assumptions. For interview preparation, prioritize fundamentals (hypothesis testing, confidence intervals, Bayesian intuition, regression assumptions, power/sample-size calculations) and practice explaining results in plain language with concrete action items. Time-boxed technical screens will probe mathematical derivations and quick probability puzzles, while onsite loops typically include deeper case-style problems and behavioral threads that test clarity and ownership. Prep with worked problems, mini-experiments you can explain, and mock interviews that force concise, business-focused answers.

18 Questions 1 Company11.04.2025
Showing 18 results
Role
Amazon logo
Amazon
Medium
Data Scientist

Compute an A/B test p-value by hand

In an A/B test on a game feature, you measure conversion rate (binary outcome). - Control: n₁ = 1000 users, x₁ = 120 conversions - Treatment: n₂ = 980...

Statistics & Math
12
0
99 people solved
Nov 4, 2025
Amazon logo
Amazon
Medium
Data Scientist

Compute p-values, CIs, and adjust multiples

Hypothesis testing and intervals in practice. Part A (z vs t): You sample n = 15 observations from a population with unknown variance and observe samp...

Statistics & Math
5
0
75 people solved
Oct 13, 2025
Amazon logo
Amazon
Hard
Data Scientist

Quantify improvement and compute required sample size

A/B Test on Spam Rate: Sample Size, Inference, and Practical Pitfalls Context: You are evaluating a new classifier that aims to reduce the spam rate (...

Statistics & Math
6
0
62 people solved
Oct 13, 2025
Amazon logo
Amazon
Hard
Data Scientist

Validate DID and IV assumptions rigorously

Causal Inference and IV: DID, TWFE, Staggered Adoption, Clustering, and 2SLS Context: You are analyzing the causal effect of a reminder on an outcome ...

Statistics & Math
8
0
100 people solved
Oct 13, 2025
Amazon logo
Amazon
Medium
Data Scientist

Compute CIs, power, and multiple testing

A/B Testing Stats: Confidence Intervals, Power, Multiple Testing, and Clustering Context: You are planning an A/B experiment on a Bernoulli outcome (c...

Statistics & Math
7
0
67 people solved
Oct 13, 2025
Amazon logo
Amazon
Hard
Data Scientist Locked

Plan and analyze an A/B test

This question evaluates expertise in experimental design and applied statistics — specifically power and sample-size calculations, clustering and desi...

Statistics & Math
5
0
66 people solved
Oct 13, 2025
Amazon logo
Amazon
Medium
Data Scientist

Identify P-Value Limitations and Complementary Approaches

Identify P-Value Limitations and Complementary Approaches A/B Testing: Limits of P-values and Better Decision Practices Scenario Your team is running ...

Statistics & Math
19
0
46 people solved
Aug 4, 2025
Amazon logo
Amazon
Medium
Data Scientist

Explain Propensity Score Matching and Assess Covariate Balance

Explain Propensity Score Matching and Assess Covariate Balance Propensity Score Matching (PSM) Context You have observational data with a binary treat...

Statistics & Math
3
0
62 people solved
Aug 4, 2025
Amazon logo
Amazon
Medium
Data Scientist

Explain Statistical Outputs to Non-Technical Stakeholders

Explain Statistical Outputs to Non-Technical Stakeholders A/B Test Dashboard Interpretation and Core Statistics Concepts Scenario You are reviewing an...

Statistics & Math
84
0
304 people solved
Aug 4, 2025
Amazon logo
Amazon
Medium
Data Scientist

Determine Probability of Both Children Being Boys

Conditional Probability: Two Children A family has two children. You learn that at least one of them is a boy. Answer the questions below and state yo...

Statistics & Math
22
0
57 people solved
Jul 12, 2025
Amazon logo
Amazon
Hard
Data Scientist

Estimate Treatment Effects Using PSM, DiD, and DML Methods

Estimate Treatment Effects Using PSM, DiD, and DML Methods Causal Impact of Marketing Campaigns: PSM, DiD, Synthetic Control, and DML Scenario You hav...

Statistics & Math
21
0
114 people solved
Aug 4, 2025
Amazon logo
Amazon
Medium
Data Scientist

Explain P-value, Confidence Interval, and Multiple Testing Adjustments

Explain P-Value, Confidence Interval, and Multiple Testing Adjustments You are running online A/B experiments to evaluate a new product launch. Assume...

Statistics & Math
45
0
153 people solved
Jul 12, 2025
Amazon logo
Amazon
Medium
Data Scientist Locked

Compute and interpret quantile loss vs RMSE

This question evaluates competency in probabilistic forecasting evaluation, including understanding of quantile (pinball) loss versus point-error metr...

Statistics & Math
10
0
76 people solved
Oct 13, 2025
Amazon logo
Amazon
Medium
Data ScientistSenior+

Calculate A/B sample size, CI, decision rules

A/B Test Design and Analysis: Signup Funnel You are designing and analyzing a two-arm A/B test for a signup funnel. Assume 1:1 traffic split and indep...

Statistics & Math
10
0
90 people solved
Oct 13, 2025
Amazon logo
Amazon
Easy
Data Scientist Locked

Answer core probability and inference questions

This question evaluates understanding of statistical inference and probability fundamentals, covering the Central Limit Theorem, p-values, Type I and ...

Statistics & Math
7
0
72 people solved
Oct 11, 2025
Amazon logo
Amazon
Medium
Data Scientist

Explain Central Limit Theorem and Its Limitations

Explain Central Limit Theorem and Its Limitations Statistics Concepts and Disease-Test Evaluation Context You are assessing core statistical concepts ...

Statistics & Math
7
0
59 people solved
Aug 4, 2025
Amazon logo
Amazon
Medium
Data ScientistSenior+

Analyze omitted-variable bias in regression

Omitted-Variable Bias, Heteroscedasticity, and Remedies Setup - True data-generating process (DGP): Y = β0 + β1·Temp + β2·Occupancy + ε - Assumption...

Statistics & Math
6
0
52 people solved
Oct 13, 2025
Amazon logo
Amazon
Hard
Data Scientist

Prove and apply statistical ML fundamentals

Technical ML/Statistics Exercises (with precise math and small computations) Assume a standard supervised learning setting with n samples, p features,...

Statistics & Math
7
0
106 people solved
Oct 13, 2025

Frequently Asked Questions

How difficult are Amazon Statistics & Math interview questions?
Amazon Statistics & Math questions range from straightforward conceptual checks to challenging applied problems. Many interviews start with probability and distribution questions that test fundamentals, then move to applied topics like hypothesis testing, confidence intervals, and regression diagnostics. Senior and specialized roles often include deeper design or derivation tasks, causal reasoning, and power/sample-size calculations. Time pressure and unclear problem statements increase perceived difficulty, and interviewers evaluate not only correct answers but reasoning, assumptions, and tradeoffs. With focused practice, candidates typically find the problems manageable, though edge cases and real-world constraints often expose gaps in preparation.
Where in the Amazon interview process does Statistics & Math appear, and what does the process look like?
Statistics & Math commonly appears across several stages of an Amazon loop: early technical screens for analytics and data roles, dedicated statistics or experimentation rounds during on-site interviews, and in role-specific interviews for data scientist, research scientist, and applied scientist positions. Expect a recruiter screen followed by a technical phone or take-home, then multiple loop interviews that include a statistics or experimentation session paired with SQL, machine learning, and behavioral questions. Interviewers assess applied intuition, the ability to design an experiment, interpret results, and communicate limitations. The focus is practical: how you would translate statistical findings into business decisions.
How should I structure my preparation timeline for Amazon Statistics & Math?
A realistic preparation timeline is four to eight weeks depending on starting level. Begin with two weeks to refresh core probability, distributions, hypothesis testing, and linear regression basics, ensuring comfort with formulas and intuition. Spend the next two weeks on applied topics like A/B test design, power calculations, metric definitions, and common diagnostics, practicing a few worked examples. Use one to two weeks for timed practice problems, coding simple statistical checks in your language of choice, and reviewing mistakes. Reserve the final week for mock interviews that simulate explanation under time constraints and polishing concise answers and assumptions.
What key subtopics within Statistics & Math should I focus on for Amazon interviews?
Focus on probability and conditional probability, discrete and continuous distributions, central limit theorem, and sampling variability. Master hypothesis testing, p-values, confidence intervals, and power/sample-size reasoning. Be fluent in A/B test design, randomization, metric selection, and common pitfalls like multiple comparisons and carryover effects. Know linear regression assumptions and diagnostics, basics of causal inference, and resampling methods such as bootstrapping. Also prepare for interpreting metrics and tradeoffs in the presence of business constraints, and for sketching simple derivations or calculations by hand when asked to justify an answer.
What are standout tips and common pitfalls to avoid when answering Amazon Statistics & Math questions?
Prioritize clear structure: state assumptions, define the null and alternative, and explain why your chosen test or metric fits the business question. Verbally justify tradeoffs and be explicit about power and sample-size considerations. Avoid overreliance on p-values; translate results into effect sizes and business impact. Common pitfalls include ignoring randomization and bias, failing to check assumptions, multiple-testing mistakes, and underpowered designs. Practice communicating succinctly under time pressure and rehearsing edge-case follow-ups; interviewers value candidates who surface limitations and next steps rather than forcing a perfect answer.

Explore more Amazon Statistics & Math interview questions

Real questions from candidate reports, grouped by role, topic and company.

By role
Other categories at Amazon
Statistics & Math questions at other companies
Browse all