Week 3 — Causal Inference I upcoming¶
July 6 – July 10, 2026 · Chapters 3.1 – 3.5 · Lab 3 · Mentor Session 3 · Problem Set 3
Correlation is cheap; the Rubin causal model makes "effect" precise, and we earn causal claims through design, not control variables.
By Friday you should be able to read a fair-lending paper, identify the threat to identification in one sentence, and say whether matching, weighting, or IV is the right knife — and what fails when it isn't.
Daily schedule¶
| Block | What |
|---|---|
| 9:00 – 12:00 | Read Ch 3.1 — Potential Outcomes, SUTVA, and the Fundamental Problem. Work nb3.1 (potential outcomes & selection bias). |
| 1:00 – 4:00 | Live problem-solving on \(\mathbb{E}[Y_1 - Y_0]\) identification. |
| 4:00 – 5:00 | Office hours |
| Block | What |
|---|---|
| 9:00 – 12:00 | Read Ch 3.2 — Matching & Propensity Scores. Work nb3.2 (PSM with balance diagnostics). |
| 1:00 – 4:00 | Read Ch 3.3 — Entropy Balancing & Doubly-Robust. Work nb3.3. |
| 4:00 – 5:00 | Office hours |
| Block | What |
|---|---|
| 9:00 – 12:00 | Read Ch 3.4 — Instrumental Variables (relevance/exclusion, 2SLS-as-FWL, LATE). Work nb3.4. |
| 2:00 – 3:30 | Mentor Session 3 — "Earning a causal claim" |
| 4:00 – 5:00 | Office hours |
| Block | What |
|---|---|
| 9:00 – 12:00 | Read Ch 3.5 — Reading IV in the Wild + Weak-IV Pathology. Work nb3.5 (conventional coverage 0.80 vs. Anderson–Rubin 0.97). |
| 1:00 – 4:00 | Lab 3 — PSM + 2SLS on HMDA (fair-lending mortgage panel). |
| 4:00 – 5:00 | Office hours |
| Block | What |
|---|---|
| 9:00 – 12:00 | Wrap PS 3.1 – 3.5. |
| 1:00 – 3:00 | Project-question shortlist — each student submits 3 candidate research questions with feasibility/novelty scores. |
| 3:00 – 4:00 | Friday demo + Week-4 preview |
| 5:00 pm | Deliverable cutoff |
What to read¶
- Ch 3.1 — Potential Outcomes & SUTVA
- Ch 3.2 — Matching & Propensity Scores
- Ch 3.3 — Entropy Balancing & Doubly-Robust
- Ch 3.4 — Instrumental Variables
- Ch 3.5 — Reading IV in the Wild + Weak-IV Pathology
Notebooks: notebooks/week-03/ — nb3.1 through nb3.5.
Lab: Lab 3 — PSM + 2SLS on HMDA.
Mentor reading: Mentor 3 — "Earning a causal claim".
What to submit (due Fri Jul 10, 5:00 pm ET)¶
- Problem Set 3 — PS 3.1 through 3.5.
- Lab 3 notebook with balance table, ATT estimate, first-stage F, and Anderson–Rubin CI.
- Project-question shortlist — three questions in
projects/<your-name>-shortlist.md, scored on feasibility and novelty. - Journal entry —
journal/week-03.md. - Pull request —
week-03 — <your name>.
Data we touch this week¶
- HMDA (Home Mortgage Disclosure Act) — fair-lending panel for Lab 3.
- CRSP and Compustat — for the PSM example.
Next week preview¶
Week 4 — Causal Inference II →. Difference-in-differences (including the staggered-adoption crisis), regression discontinuity, synthetic control, and shift-share designs.