PracHub
QuestionsPremiumCoachesLearningGuidesInterview Prep
|Home/Statistics & Math/Meta

Estimate first selection round with/without replacement

Last updated: Mar 29, 2026

Quick Overview

This question evaluates understanding of probability and expectation concepts, specifically sampling with and without replacement, per-round success probabilities, geometric waiting-time distributions, and first-order approximation methods in a finite-population context.

  • medium
  • Meta
  • Statistics & Math
  • Data Scientist

Estimate first selection round with/without replacement

Company: Meta

Role: Data Scientist

Category: Statistics & Math

Difficulty: medium

Interview Round: Onsite

There are 1000 people. Each round, 10 selections are made. Case 1 (without replacement across the entire process): in total 100 rounds are run to select everyone exactly once (10 per round); a given person's position in the random permutation determines their round. For a fixed person, compute the expected round index of their first selection. Case 2 (with replacement within and across rounds): each round consists of 10 independent draws with replacement from the 1000 people; a person is counted as 'selected' in a round if they are drawn at least once that round. For a fixed person, derive exactly the per-round selection probability p and the expected number of rounds until their first selection (a geometric random variable). Provide a first-order approximation to p.

Quick Answer: This question evaluates understanding of probability and expectation concepts, specifically sampling with and without replacement, per-round success probabilities, geometric waiting-time distributions, and first-order approximation methods in a finite-population context.

Related Interview Questions

  • Compute probability an account is fake - Meta (easy)
  • Compute Bayes probability for fake accounts - Meta (easy)
  • Compute probabilities for chatbot response quality - Meta (easy)
  • Compute posterior fake probability using Bayes' rule - Meta (medium)
  • Estimate bots and CI from DAU spike - Meta (medium)
Meta logo
Meta
Oct 13, 2025, 9:49 PM
Data Scientist
Onsite
Statistics & Math
1
0
Loading...

Probability of First Selection Across Rounds

Context: There are 1,000 people. Each round, 10 draws are made. Consider a fixed person.

Case 1: Without Replacement Across the Entire Process

  • The process is a single random permutation of all 1,000 people.
  • People are partitioned into 100 rounds, 10 per round (positions 1–10 → round 1, 11–20 → round 2, …, 991–1000 → round 100).
  • Question: What is the expected round index of the person's (only) selection?

Case 2: With Replacement Within and Across Rounds

  • Each round consists of 10 independent draws with replacement from the 1,000 people.
  • A person is counted as "selected" in a round if they are drawn at least once in that round (multiple hits in the same round still count as one selection for that round).
  • Tasks:
    1. Derive the exact per-round selection probability p for the fixed person.
    2. Assuming rounds are i.i.d. with success probability p, the number of rounds until first selection is geometric. Compute its expected value.
    3. Provide a first-order approximation to p.

Solution

Show

Submit Your Answer to Earn 20XP

Sign in to leave a comment

Loading comments...

Browse More Questions

More Statistics & Math•More Meta•More Data Scientist•Meta Data Scientist•Meta Statistics & Math•Data Scientist Statistics & Math
PracHub

Master your tech interviews with 8,000+ real questions from top companies.

Product

  • Questions
  • Learning Tracks
  • Interview Guides
  • Resources
  • Premium
  • For Universities
  • Student Access

Browse

  • By Company
  • By Role
  • By Category
  • Topic Hubs
  • SQL Questions
  • Compare Platforms
  • Discord Community

Support

  • support@prachub.com
  • (916) 541-4762

Legal

  • Privacy Policy
  • Terms of Service
  • About Us

© 2026 PracHub. All rights reserved.