causal inference · Live seminars
Causal Inference for Policy
Potential outcomes, DAG discipline, and when difference-in-differences is defensible in policy memos.
ZAR 2 100 · 8 weeks
Software: Stata · Method focus: causal · Level: advanced · Assessment: policy memo
Description
You map assumptions onto estimands, sketch selection stories, and practice writing limitation paragraphs that reviewers accept. Includes worked examples from public employment programmes.
What is included
- DAG exercises with mentor feedback
- DiD table templates with parallel trends notes
- Short policy memo with peer critique
- Office hours on IV relevance and exclusion
- Reading list on local average treatment effects
- Checklist for external validity claims
Outcomes
- Match estimand to estimator in a two-page brief
- State identifying assumptions in plain language
- Flag threats to validity before running new specs
Prof. Amir Hassan
Applied microeconomist focusing on programme evaluation in Southern Africa.
Reviews
The DiD memo template saved us a full drafting cycle — especially the parallel trends paragraph our director wanted tightened.
Questions
We reference randomization inference but do not run field RCTs in this cohort.
Yes, with ethics clearance documentation if human subjects are involved.
Synthetic control and matrix completion are out of scope for this version.