Statistics & Math Interview Questions

Showing 31–60 of 124 results

• DoorDashHard·Data Scientist
  Define and compute retention and churn precisely

  This question evaluates a Data Scientist's competency in statistical measurement of user retention and churn, covering cohort definition, activity rules, risk-set-aware retention and churn formulas, censoring and delayed conversion handling, and survival-analysis concepts like time-to-churn, hazard, and cumulative incidence.

  Statistics & MathOnsite
• GleanMedium·Data Scientist
  Expectation and Variance of a Sum of Two Random Variables, Including Covariance

  Asks how to compute the expectation and variance of the sum of two random variables, in general and when they are independent. Tests derivation from definitions, the role of covariance, generalizing to weighted sums and differences, and applying the result to data-science problems such as comparing group means.

  Statistics & MathOnsite
• GleanEasy·Data Scientist
  What are sum expectation and variance?

  This question evaluates understanding of expectation, variance and covariance and the ability to reason about how sums of random variables affect their moments, including the role of covariance and the special case of independence in the Statistics & Math domain.

  Statistics & MathOnsite
• GoogleMedium·Data Scientist
  Can bootstrap help reduce variance

  This question evaluates understanding of bootstrap resampling, the distinction between estimating an estimator’s variance and methods that actually reduce estimator variance, and related ensemble reasoning for a Data Scientist role in the Statistics & Math domain.

  Statistics & MathOnsite
• GoogleHard·Data Scientist
  Design human review to estimate model accuracy

  This interview question evaluates statistical assumptions, formulas, estimation strategy, uncertainty, edge cases, and interpretation in a realistic interview setting. A strong answer for Design human review to estimate model accuracy states assumptions, handles edge cases, explains trade-offs, and shows how to validate the result clearly.

  Statistics & MathOnsite
• GoogleHard·Data Scientist
  Estimate Population Mean and Conversion Rate Accurately

  Evaluates core statistical inference skills across p-values, confidence intervals, standard error reduction, tail probability estimation, efficient conversion-rate sampling, and truncated-normal estimation. Strong answers state assumptions, use appropriate intervals and weights, and avoid common interpretation errors.

  Statistics & MathOnsite
• GoogleMedium·Data Scientist
  Explain P-values and Confidence Intervals in Simple Linear Regression

  Review p-values, confidence intervals and slope inference in simple linear regression, including assumptions and causal limits.

  Statistics & MathOnsite
• GoogleEasy·Data Scientist
  Explain mixed models and fixed vs random effects

  This question evaluates understanding of mixed effects models, the conceptual distinction between fixed and random effects, interpretation of coefficients and variance components, and awareness of practical pitfalls when modeling hierarchical or panel data.

  Statistics & MathOnsite
• GoogleMedium·Machine Learning Engineer
  Generate values by weighted probabilities

  This question evaluates understanding of probabilistic sampling, numerical stability, input validation, and algorithmic trade-offs for streaming weighted random draws, including generator determinism via seeding and methods such as CDF/binary-search versus alias for performance.

  Statistics & MathOnsite
• GoogleMedium·Data Scientist
  Interpreting a Dosage-Recovery Regression: P-values, CIs, Causality

  Given a fitted dosage-vs-recovery-time regression, explain exactly what the reported p-value and confidence interval mean, what assumptions back that inference, how a confidence interval differs from a prediction interval, and why the result alone cannot establish causality. Tests statistical precision under pressure.

  Statistics & MathOnsitestatisticslinear-regressionhypothesis-testing
• GoogleMedium·Data Scientist
  Reading a Regression Slope: P-values, Intervals, and Causality

  Use a fitted slope from a one-predictor regression to explain precisely what a p-value and a confidence interval mean, what assumptions they rest on, how confidence and prediction intervals differ, and why significance is not causation. Tests statistical precision under interview pressure.

  Statistics & MathOnsitestatisticslinear-regressionhypothesis-testing
• GoogleMedium·Data Scientist
  Understand Simpson's Paradox with Simple Examples

  This interview question evaluates statistical assumptions, formulas, estimation strategy, uncertainty, edge cases, and interpretation in a realistic interview setting. A strong answer for Understand Simpson's Paradox with Simple Examples states assumptions, handles edge cases, explains trade-offs, and shows how to validate the result clearly.

  Statistics & MathOnsite
• InstacartHard·Data Scientist
  Choose tests under non‑normal, unequal variance

  This question tests the ability to select and justify statistical inference methods for metrics that violate standard parametric assumptions, specifically when dealing with heavy-tailed, heteroskedastic, and zero-inflated distributions.

  Statistics & MathOnsiteStatisticsA/B TestingInference
• LiftoffHard·Machine Learning Engineer
  Probability That the Sum of 1,000 Fair Die Rolls Exceeds a Threshold

  A fair six-sided die is rolled 1,000 times; estimate the probability that the sum of the faces exceeds a threshold, including a threshold of 600. Tests the mean and variance of one roll, how they scale for a sum of independent rolls, the normal approximation with a continuity correction, and range checks.

  Statistics & MathOnsite
• LinkedInMedium·Data Scientist
  Estimate an Experiment’s Mean Effect from Cohort and Metric Tables

  Estimate experimental impact from cohort and timestamped metric data using user-level outcomes, difference in means, confidence intervals, and valid testing.

  Statistics & MathOnsite
• LinkedInHard·Data Scientist
  Measure Causal Impact of Self-Selected App Redesign

  LinkedIn causal inference prompt on measuring a self-selected app redesign, covering potential outcomes, ATT, propensity matching or weighting, staggered difference-in-differences, event studies, covariates, balance, and pre-trend validation.

  Statistics & MathOnsite
• LyftMedium·Data Scientist
  Compute Poisson supply–demand match probability

  This question evaluates understanding of probability distributions and stochastic modeling—specifically Poisson processes, the Skellam distribution for differences, expectation of minima, and probabilistic capacity planning within the Statistics & Math domain.

  Statistics & MathOnsite
• MetaMedium·Data Scientist
  Analyze User Comment Distribution and Sampling Effects

  Evaluates statistical reasoning about right-skewed user comment counts and sampling distributions. Strong answers place median, mean, and p95 correctly, then use the Central Limit Theorem to explain why group averages become less variable and more symmetric.

  Statistics & MathOnsite
• MetaMedium·Data Scientist
  Analyze User Transfer Distribution in Initial Launch Period

  Evaluates distribution analysis for transfer counts after a new P2P payments launch. Strong answers specify the denominator, describe zero inflation and right skew, locate mean, median, mode, and p95, and explain how adoption, cohorts, and exposure change the distribution over time.

  Statistics & MathOnsite
• MetaMedium·Data Scientist
  Analyze skewed comments and sampling effects

  This question evaluates understanding of descriptive statistics for right‑skewed count data and inferential concepts such as sampling distributions, standard error, percentiles, and the Central Limit Theorem, and is commonly asked to probe how skewness and sample size affect central tendency measures and uncertainty in estimates.

  Statistics & MathOnsite
• MetaMedium·Data Scientist
  Annotating and forecasting a long‑tail distribution

  Evaluates robust summarization and forecasting for long-tailed daily share-count distributions. Strong answers compute mean, median, P1, P99, visualize with log scales or percentile bands, and forecast seasonality with heavy-tail-aware methods.

  Statistics & MathOnsite
• MetaHard·Data Scientist
  Apply sequential testing without p-hacking

  This question evaluates a candidate's competency in sequential hypothesis testing, alpha‑spending rules (Pocock vs O’Brien–Fleming), always‑valid inference methods such as mixture‑SPRT and e‑values, and strategies for controlling error rates across multiple metrics.

  Statistics & MathOnsite
• MetaMedium·Data Scientist
  Calculate Ad Insertion Statistics for Two Methods

  Meta statistics prompt comparing two ad-insertion strategies, covering binomial moments, deterministic quotas, tail probability for excessive ads, adjacent ad pairs, and risk of long ad runs.

  Statistics & MathOnsite
• MetaEasy·Data Scientist
  Calculate Expected Impressions and Probability for Users

  Meta probability question on random ad-impression allocation, including binomial expectation, probability a user sees at least one impression, expected unique reach, and large-population approximations.

  Statistics & MathOnsite
• MetaEasy·Data Scientist
  Calculate Posterior Fraud Probability Using Bayes' Theorem

  Calculate Posterior Fraud Probability Using Bayes' Theorem evaluates statistical assumptions, formulas, estimation strategy, uncertainty, edge cases, and interpretation in a realistic interview setting. A strong answer states assumptions, handles edge cases, explains trade-offs, and shows how to validate the result clearly.

  Statistics & MathOnsite
• MetaEasy·Data Scientist
  Calculate Posterior Probability of Flagged User Being Bad Actor

  Evaluates Bayesian inference and confusion-matrix reasoning for abuse detection. Strong answers derive posterior precision from prevalence, TPR, and FPR, define Type I and Type II errors, and compute true positives and false positives for 10 million users.

  Statistics & MathOnsite
• MetaEasy·Data Scientist
  Calculate Probability of Honest and Relevant Chatbot Answers

  Calculate Probability of Honest and Relevant Chatbot Answers evaluates statistical assumptions, formulas, estimation strategy, uncertainty, edge cases, and interpretation in a realistic interview setting. A strong answer states assumptions, handles edge cases, explains trade-offs, and shows how to validate the result clearly.

  Statistics & MathOnsite
• MetaMedium·Data Scientist
  Characterize and compare transfer-count distributions over time

  This question evaluates a candidate's ability to model zero-inflated, heavy-tailed count data, interpret distributional summaries and percentiles, reason about temporal shifts from retention and fraud dynamics, and identify robust summary statistics and diagnostics for a Data Scientist role.

  Statistics & MathOnsite
• MetaMedium·Data Scientist
  Characterize metric distribution and quantiles

  This question evaluates a data scientist's competency in descriptive statistics and robust metric analysis within the Statistics & Math domain, focusing on distribution characterization, empirical quantiles, measures of central tendency, and the effects of outlier handling on variance and experimental power.

  Statistics & MathOnsite
• MetaEasy·Data Scientist
  Compare Ad-Insertion Strategies: Expected Ads and Probabilities

  Evaluates probability, expectation, and variance for newsfeed ad-insertion strategies. Strong answers compute expected ad load, at-least-one exposure probability, and explain why equal expected counts can still produce different user experiences.

  Statistics & MathOnsite