Statistics & Math interview questions

Showing 301–330 of 439 results

• Estimate billboard reach and impressionsHard

  This question evaluates probabilistic modeling and statistical inference skills applied to audience measurement and attribution, covering Poisson-based frequency modeling, visibility-adjusted impressions, panel-based deduplication and expansion, uncertainty propagation (delta method/bootstrap), and Markov-chain attribution within the Statistics & Math / Data Science domain. It is commonly asked to test the ability to convert traffic and visibility inputs into quantitative reach and impressions with propagated confidence intervals, reason about attribution via a 3-state Markov chain, identify measurement biases, and demonstrates both conceptual understanding and practical application through numerical estimation and sensitivity analysis.

  PinterestData ScientistStatistics & MathOnsite
• Implement Bootstrap and Jackknife Estimates on a DataFrameMedium

  Tests practical statistical computing skills by requiring the implementation of bootstrap confidence intervals and jackknife variance estimation for arbitrary metrics evaluated on pandas DataFrames.

  PinterestData ScientistStatistics & MathOnsite
• Calibrate the Risk-Aversion ParameterMedium

  Calibrate mean-variance risk aversion to target risk and out-of-sample utility while accounting for units, scaling, constraints, and regimes.

  Point72Quantitative ResearcherStatistics & MathOnline Assessment
• Choose Practical Portfolio ConstraintsMedium

  Specify practical stat-arb constraints for neutrality, exposure, concentration, turnover, liquidity, and borrow, including feasibility checks.

  Point72Quantitative ResearcherStatistics & MathOnline Assessment
• Derive Annualized Portfolio VolatilityMedium

  Derive portfolio volatility from factor covariance and stock-specific variance, then annualize it with consistent horizons and units.

  Point72Quantitative ResearcherStatistics & MathOnline Assessment
• Derive the Optimal Bid in a Two-Bidder Second-Price AuctionHard

  Two risk-neutral bidders compete for one item in a sealed-bid second-price auction. Explain the assumptions and derivation clearly, check edge cases, and show how the result changes when those assumptions no longer hold.

  Point72Quantitative ResearcherStatistics & MathTechnical Screen
• Estimate Factor Returns and Residual ReturnsMedium

  Derive weighted factor-return estimates from a sample, then compute residual returns across a larger trading universe with stable linear algebra.

  Point72Quantitative ResearcherStatistics & MathOnline Assessment
• Find the Two Heaviest Balls with Minimum ComparisonsHard

  There are 32 balls with distinct weights. Explain the assumptions and derivation clearly, check edge cases, and show how the result changes when those assumptions no longer hold.

  Point72Quantitative ResearcherStatistics & MathTechnical Screen
• Formulate a Single-Period Stat-Arb ObjectiveMedium

  Formulate a convex single-period statistical-arbitrage objective balancing expected return against covariance risk, with consistent units.

  Point72Quantitative ResearcherStatistics & MathOnline Assessment
• Linearize an L1 Trading PenaltyMedium

  Linearize weighted absolute trading costs with auxiliary variables, complete LP constraints, the correct objective sign, and an equivalence proof.

  Point72Quantitative ResearcherStatistics & MathOnline Assessment
• Model Trading Frictions in Portfolio OptimizationMedium

  Extend a stat-arb objective with unit-consistent commissions, spread, slippage, market impact, turnover, borrow, and concentration costs.

  Point72Quantitative ResearcherStatistics & MathOnline Assessment
• Price a coin-doubling game rationallyHard

  This question evaluates understanding of probability distributions, divergent expected values, utility theory, and optimal betting strategies (Kelly staking) within the Statistics & Math domain for a Data Scientist role.

  Point72Data ScientistStatistics & MathOnline Assessment
• Prove That Lasso Can Be Written as a Quadratic ProgramMedium

  Derive the Lasso quadratic-program formulation and prove equivalence, convexity, minimizer correspondence, and rank-related uniqueness.

  Point72Quantitative ResearcherStatistics & MathTechnical Screen
• Solve Three Probability ModelsMedium

  Solve three independent probability questions. Explain the assumptions and derivation clearly, check edge cases, and show how the result changes when those assumptions no longer hold.

  Point72Quantitative ResearcherStatistics & MathTechnical Screen
• Justify synthetic control and handle inferenceHard

  This question evaluates understanding of synthetic control methodology, causal identification assumptions, estimation and regularization choices, inference techniques for pointwise and cumulative effects, and diagnostic checks for panel time-series interventions.

  RedditData ScientistStatistics & MathTechnical Screen
• Apply Bayes' rule to compute posteriorHard

  Evaluates understanding of Bayes’ rule and basic probability concepts—specifically computing a posterior probability P(A|B) and the marginal likelihood P(B) from a prior and likelihoods—in the Statistics & Math domain at a foundational-to-intermediate abstraction level.

  RobloxData ScientistStatistics & MathOnline Assessment
• Compute A/B test sample size and Bayes posteriorEasy

  This question tests the ability to calculate required sample size for a frequentist A/B test using power analysis and to perform a Bayesian update for a Bernoulli process.

  RobloxData ScientistStatistics & MathOnline Assessment
• Compute sample size and Bayes posteriorHard

  This question evaluates understanding of statistical inference and probabilistic reasoning by combining a frequentist power/sample-size computation for a two-sided two-sample z-test with a basic Bayesian conditional probability calculation.

  RobloxData ScientistStatistics & MathOnline Assessment
• Derive variance and CTR confidence intervalsMedium

  This problem tests your understanding of statistical inference for binomial proportions, specifically calculating confidence intervals for Click-Through Rates (CTR) and understanding how to aggregate variability across multiple days.

  RobloxData ScientistStatistics & MathTechnical Screen
• Implement robust one/two-sided p-value functionHard

  This question evaluates proficiency in statistical hypothesis testing, numerical computing, and numerical stability when implementing p-value calculations for z and t distributions, and it targets the Statistics & Math domain for a data scientist role.

  RobloxData ScientistStatistics & MathTechnical Screen
• Explain Bias-Variance Trade-off Simply for StakeholdersEasy

  Evaluates the ability to explain the bias-variance trade-off to non-technical stakeholders. Strong answers define bias as underfitting, variance as overfitting, use a clear analogy, and explain why models must generalize to new data rather than memorize training data.

  RokuData ScientistStatistics & MathTechnical Screen
• Analyze Knights-and-Knaves Dialogue VariantsMedium

  Analyze Knights-and-Knaves Dialogue Variants 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.

  SIG (Susquehanna)Data ScientistStatistics & MathTechnical Screen
• Bayesian Posterior That k Red Parts Came From Factory AMedium

  A probability question on Bayesian updating: one of two factories with different red-part rates is chosen at random, and k parts drawn from it are all red. It asks for the conditional probability that the parts came from the first factory, testing likelihoods, priors and careful evaluation of powers.

  SIG (Susquehanna)Quantitative TraderStatistics & MathOnline Assessment
• Calculate Expected Payoff from Two DiceEasy

  This problem tests fundamental knowledge of probability, expected value, and partitioning sample spaces to calculate the average outcome of a stochastic process.

  SIG (Susquehanna)Data ScientistStatistics & MathOnline Assessment
• Compute Conditional Age Probability Among CyclistsEasy

  This question tests the ability to interpret a contingency table and calculate conditional probabilities by narrowing the sample space to a specific sub-group.

  SIG (Susquehanna)Data ScientistStatistics & MathOnline Assessment
• Compute answers to probability and counting puzzlesHard

  This problem tests your ability to model sequential probabilistic events using combinatorics and conditional probability, specifically requiring the calculation of expected outcomes in a game of chance.

  SIG (Susquehanna)Data ScientistStatistics & MathTechnical Screen
• Compute ruin probability with bold bettingMedium

  This question evaluates understanding of probability theory and stochastic processes, focusing on ruin probabilities under a bold betting strategy with state-dependent wagers.

  SIG (Susquehanna)Data ScientistStatistics & MathOnline Assessment
• Compute stockout sufficiency probabilityEasy

  This question evaluates understanding of discrete probability and binomial models for independent Bernoulli trials, testing the ability to compute the probability that a limited resource meets random demand.

  SIG (Susquehanna)Data ScientistStatistics & MathOnline Assessment
• Compute the Probability Two Bloom Intervals OverlapEasy

  Calculate the probability that two independent events with different known durations overlap within a fixed time window using geometric probability.

  SIG (Susquehanna)Data ScientistStatistics & MathOnline Assessment
• Condition on a Chosen Rabbit LosingEasy

  Calculate the conditional probability that a randomly selected rabbit is the least athletic of three rabbits given that it lost its benchmark race, using Bayes' theorem.

  SIG (Susquehanna)Data ScientistStatistics & MathOnline Assessment