PracHub
QuestionsPremiumLearningGuidesInterview PrepNEWCoaches
|Home/Statistics & Math/Meta

Contrast OLS, DiD, and PSM assumptions

Last updated: Mar 29, 2026

Quick Overview

This question evaluates proficiency in causal inference and econometric methods, specifically the ability to contrast OLS, two-way fixed-effects DiD, and propensity score matching (PSM) assumptions, diagnostics, and estimation strategies for treatment-effect analysis.

  • Medium
  • Meta
  • Statistics & Math
  • Data Scientist

Contrast OLS, DiD, and PSM assumptions

Company: Meta

Role: Data Scientist

Category: Statistics & Math

Difficulty: Medium

Interview Round: Technical Screen

For the shuttle impact problem, contrast OLS, DiD, and PSM rigorously. Do the following: 1) write the OLS and two-way fixed-effects DiD regression equations with all variables and fixed effects defined; 2) state the identifying assumptions for each (exogeneity, SUTVA, parallel trends, overlap/positivity) and the specific violations you expect here; 3) propose and justify a placebo test and a balance test, including metrics (e.g., standardized mean differences) and acceptance thresholds; 4) explain matching with vs. without replacement and the bias–variance–computational trade-offs; 5) after matching, show exactly how you would estimate effects (PSM+DiD) and compute ATT with valid standard errors; 6) discuss consequences of limited overlap and how you would diagnose and remedy it (trimming, calipers, alternative link).

Quick Answer: This question evaluates proficiency in causal inference and econometric methods, specifically the ability to contrast OLS, two-way fixed-effects DiD, and propensity score matching (PSM) assumptions, diagnostics, and estimation strategies for treatment-effect analysis.

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
Technical Screen
Statistics & Math
1
0
Loading...

For the shuttle impact problem, contrast OLS, DiD, and PSM rigorously. Do the following: 1) write the OLS and two-way fixed-effects DiD regression equations with all variables and fixed effects defined; 2) state the identifying assumptions for each (exogeneity, SUTVA, parallel trends, overlap/positivity) and the specific violations you expect here; 3) propose and justify a placebo test and a balance test, including metrics (e.g., standardized mean differences) and acceptance thresholds; 4) explain matching with vs. without replacement and the bias–variance–computational trade-offs; 5) after matching, show exactly how you would estimate effects (PSM+DiD) and compute ATT with valid standard errors; 6) discuss consequences of limited overlap and how you would diagnose and remedy it (trimming, calipers, alternative link).

Comments (0)

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 7,500+ 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.