This lecture connects Difference-in-Differences to timing, comparison groups, and design diagnostics that shape the causal interpretation.
Quasi-Experiments and Natural Experiments
Quasi-Experiments
Natural Experiments
Lecture Notes
This course studies causal designs that draw credibility from timing, thresholds, policy rules, encouragements, shocks, or comparison units. These designs are common in product, policy, education, marketing, public-sector, and platform settings where randomized evidence is unavailable or incomplete.
The objective is to build design-first judgment. By the end of the course, a reader should be able to explain what source of variation identifies the effect, diagnose the key assumptions, distinguish pre-trend evidence from proof, reason about treatment timing and spillovers, and write a careful decision interpretation.
Lecture Sequence
This lecture develops Event Studies with examples that make assumptions, diagnostics, and interpretation visible.
03. Staggered Adoption and Modern DiD Issues
This lecture uses Staggered Adoption and Modern DiD Issues to clarify the analyst’s question, evidence, assumptions, and decision implications.
This lecture studies synthetic control through donor-pool choice, pre-period fit, placebo checks, and decision interpretation.
This lecture frames Regression Discontinuity as a decision problem and asks what evidence can be trusted, challenged, and communicated.
This lecture builds intuition for Instrumental Variables and ties the result to model choice, uncertainty, and action.
This lecture applies Encouragement Designs with emphasis on diagnostics, tradeoffs, and evidence limits.
This lecture develops Interrupted Time Series as a practical pattern for analysis, diagnostics, and decision support.