48 lessons
Statistics · Module 3Module material

Probability interview questions

26 questions interviewers ask on this module’s topics, from first principles to the follow-ups. Each comes with a model answer.

Events and probability

  1. What is a sample space, an event and an empirical probability?●
  2. How do you get the probability of "A or B" from counts?●
  3. What is the complement of Amount > 100, and why does the boundary matter?●
  4. An amount rule captures 26.4% of labelled frauds. Is it a good fraud screen?●●
  5. What does the law of large numbers say, and what does it not say?●●

Conditional probability and independence

  1. What is a conditional probability?●
  2. Why are P(spam | flagged) and P(flagged | spam) different?●
  3. How do you test whether two events are independent?●●
  4. Are mutually exclusive events independent?●●
  5. What does Naive Bayes assume about independence, and how would you check it?●●

Bayes' theorem

  1. State Bayes' theorem in words.●
  2. A test catches 90% of positives with 10% false positives. Why can most of its positives be false?●●
  3. What is the law of total probability, and where does it appear in Bayes' theorem?●●
  4. Which improves a detector's posterior more at a 1% prior: fewer false alarms or a higher catch rate?●●
  5. Three separate flags fire. How do you update, and when is the update wrong?●●

Random variables, expectation and variance

  1. What is a random variable?●
  2. What is an expected value, and how is it computed from a distribution?●
  3. How would you choose between two decision rules with different error types?●●
  4. The cost matrix is uncertain. How do you report the decision?●●
  5. Why report a variance or standard deviation with an expected cost?●●
  6. What is the difference between a discrete and a continuous random variable?●●

Monte Carlo simulation

  1. What is a Monte Carlo simulation?●
  2. How do you check that a simulation is working before trusting it?●●
  3. Why evaluate every candidate on the same simulated periods?●●
  4. What is the difference between Monte Carlo error and assumption uncertainty?●●
  5. A simulation input contains one extreme recorded day. What do you do?●●

Directory listings

  • SchoolWhool on Siteefy