Session 10: Before the session

Did the policy do anything, or were the groups different to begin with?

Thierry Warin, PhD

HEC Montréal

Outline · Before class

  1. What to do, and how long it takes
  2. Read the paper
  3. Get BOTH datasets
  4. The course data
  5. Self-check, and what the session will ask of you
  6. Before you run anything

What session 10 is for

The mathematics

  • State the fundamental problem of causal inference, and write the naive difference as effect plus selection bias
  • Explain why randomisation solves it, and what it costs
  • Estimate a propensity score and match treated to control units
  • Report an ATT, and state the assumption it rests on

The Python

  • Simulate both potential outcomes and watch a coin flip recover the effect
  • Fit a propensity score with smf.logit and match with NearestNeighbors
  • Check overlap and balance (standardised differences) before reading any estimate
  • Score the naive, regression and matched estimates against an experimental benchmark

The international business

  • Estimate the effect of a programme that firms chose rather than drew
  • Say why the firms that joined were already different
  • Report how far the naive comparison would have been wrong

Before the session

Did the policy do anything, or were the groups different to begin with?

What to do, and how long it takes

1 Read the paper 45–60 min
2 Get both datasets — the paper’s, and your own angle 10 min
3 Attempt the self-check 15 min

The lecture assumes all three are done. The practice assumes the data is already cached on your machine.

1 · Read the paper

Atkin, Khandelwal & Osman (2017), Exporting and Firm Performance: Evidence from a Randomized Experiment, Quarterly Journal of Economics 132(2)

Read the article

Read for the argument, not for coverage. Annotate four things:

1 The research question, in one sentence
2 The method or design, and why they chose it
3 The strongest single piece of evidence
4 One limitation you would raise as a referee

2 · Get BOTH datasets

The practice uses two: the paper’s data, to reproduce one of its results, and your own angle, to apply the method. Have both before you arrive.

a) The paper’s replication package — Harvard Dataverse

10.7910/DVN/QOGMVI

Download it from the terminal — no Dataverse account, no browser. From the repository root:

macOS and Linux

mkdir -p 10-causal-inference-foundations/data/replication && cd 10-causal-inference-foundations/data/replication
curl -L -o replication.zip \
  "https://dataverse.harvard.edu/api/access/dataset/:persistentId/?persistentId=doi:10.7910/DVN/QOGMVI"
unzip -q replication.zip && rm replication.zip && cd -

Windows PowerShell

mkdir 10-causal-inference-foundations/data/replication; cd 10-causal-inference-foundations/data/replication
curl.exe -L -o replication.zip `
  "https://dataverse.harvard.edu/api/access/dataset/:persistentId/?persistentId=doi:10.7910/DVN/QOGMVI"
Expand-Archive replication.zip .; Remove-Item replication.zip; cd -

data/replication/ is git-ignored, so nothing large is committed. Keep the authors’ own structure — do not tidy it.

This deposit sits behind a Dataverse guestbook, so the curl recipe above returns an error for this session. Open the DOI in a browser, click Access Dataset -> Download ZIP, accept the terms once, and unzip it to the same folder. Everything after that is identical.

2b · The course data

import qmib

core = qmib.load("core")                # the spine, shared by every group
mine = qmib.load("angle_c_country")     # REPLACE with your own angle

print(core.shape, mine.shape)

Run this before you arrive. It downloads once and caches as parquet, so the practice works whatever the room’s wifi is doing.

angle_c_country is an example, not your angle. Yours is named in RESEARCH-MANDATES.md.

The spine’s documented treatment, with a known effect to recover · everything available: qmib.catalog()

3 · Self-check, and what the session will ask of you

Answer these on paper. If you cannot, that is what the lecture is for.

  1. In one sentence: what does this session’s method let you claim that the previous one did not?
  2. What must be true of your data for it to apply?
  3. Which claim in the paper rests on this method — and how hard does it lean on it?

By the end of the session you should be able to:

  • State the fundamental problem of causal inference
  • Explain why randomisation solves it, and what it costs
  • Estimate a propensity score and use it to match treated to control units
  • Check overlap and balance, and say what to do when they fail
  • Report an ATT, and state the assumption it rests on

Arrive with: the paper read and annotated · the replication package unzipped · qmib.load() run at least once · your self-check answers on paper.

Before you run anything

Everything runs in the VS Codium terminal, from the repository root, with the project environment active. Activate it in every new terminal, not only the first:

source .venv/bin/activate        # macOS and Linux
.venv\Scripts\activate           # Windows (PowerShell)

Your prompt must begin with (.venv). Without it python and pip are your system’s, and nothing the course installed is visible to them.

ModuleNotFoundError means the packages went somewhere else, or were never installed. Check the prompt first, then pip install -r requirements.txt — from the repository root, with the environment active. Building the environment from scratch is Session 01, step 1.4.

Session 10: README.md · the lecture: 01-lecture/