How do I approach Analytics & Experimentation interview questions?

Analytics & Experimentation questions require understanding of core concepts and practice. PracHub provides solutions with explanations to help you master analytics & experimentation interviews.

What difficulty level is this interview question?

This is a hard difficulty Analytics & Experimentation question, commonly asked during Onsite rounds at Freddie Mac.

What role is this question designed for?

This question is commonly asked for Data Scientist candidates at Freddie Mac during technical interviews.

Design a quasi-experiment for FHFA policy | Freddie Mac Interview Question

Quick Overview

This question evaluates causal inference and quasi-experimental study design skills, including identification strategy, diagnostics and robustness checks, handling spillovers and partial compliance, outcome measurement, and inference on loan-level financial data.

Causal Impact of FHFA's April 15, 2024 Multifamily Underwriting Tightening

Context

On April 15, 2024, FHFA tightened underwriting standards for multifamily loans purchased/guaranteed by the GSEs. You are asked to design a quasi-experimental study to estimate the policy's causal impact on:

Monthly originations (counts and dollars), and
12‑month delinquency on newly originated loans.

Assume you have loan-level GSE data with application/lock/close dates, underwriting metrics (LTV, DSCR, debt yield), property characteristics, geography (MSA), lender, and performance.

Tasks

Design choices and identification
- Define treatment, control, unit of analysis (MSA or property_type×MSA), and timing.
- Justify a staggered difference‑in‑differences (DiD) with Sun–Abraham event‑study estimators versus synthetic control.
Diagnostics and robustness
- Describe how you will test and visualize pre‑trends.
- Specify falsification tests (placebo policy dates, treatment leads) and robustness checks (alternative time windows, event‑study binning).
Spillovers, composition, and interference
- Address capital reallocation spillovers across MSAs, compositional shifts in borrowers/deals, and interference via lender networks.
- Propose instrumentation or reweighting if needed.
Measurement and inference
- Specify outcome definitions precisely, the clustering level for standard errors, and how you will communicate uncertainty to executives (e.g., prediction intervals vs. standard errors/confidence intervals).
Data and compliance
- Outline data requirements and how you will handle policy anticipation (pull‑forward) and partial compliance (grandfathered pipeline, waivers).

Quick Overview

Context

Monthly originations (counts and dollars), and

12‑month delinquency on newly originated loans.

Assume you have loan-level GSE data with application/lock/close dates, underwriting metrics (LTV, DSCR, debt yield), property characteristics, geography (MSA), lender, and performance.

Tasks

Design choices and identification

Define treatment, control, unit of analysis (MSA or property_type×MSA), and timing.
Justify a staggered difference‑in‑differences (DiD) with Sun–Abraham event‑study estimators versus synthetic control.

Diagnostics and robustness

Describe how you will test and visualize pre‑trends.
Specify falsification tests (placebo policy dates, treatment leads) and robustness checks (alternative time windows, event‑study binning).

Spillovers, composition, and interference

Address capital reallocation spillovers across MSAs, compositional shifts in borrowers/deals, and interference via lender networks.
Propose instrumentation or reweighting if needed.

Measurement and inference

Specify outcome definitions precisely, the clustering level for standard errors, and how you will communicate uncertainty to executives (e.g., prediction intervals vs. standard errors/confidence intervals).

Data and compliance

Outline data requirements and how you will handle policy anticipation (pull‑forward) and partial compliance (grandfathered pipeline, waivers).

Design a quasi-experiment for FHFA policy

Quick Overview

Causal Impact of FHFA's April 15, 2024 Multifamily Underwriting Tightening

Context

Tasks

Solution

Submit Your Answer to Earn 20XP

Design a quasi-experiment for FHFA policy

Quick Overview

Causal Impact of FHFA's April 15, 2024 Multifamily Underwriting Tightening

Context

Tasks

Solution

Submit Your Answer to Earn 20XP