Instrumental Variables Lab — When OLS Lies
Three real published cases — schooling, the Vietnam draft, colonial institutions — walked end to end.
About this simulation
A four-round empirical-econometrics workshop set in the Costaria National Productivity Council. Participants must deliver a defensible estimate of the PNCG management-training program's productivity effect on a panel of 1,500 SMEs over 2018–2024.
Across four rounds — naive pooled OLS, omitted-variable bias decomposition, fixed-effects (one-way, two-way, random) with the Hausman test, and robustness (cluster-robust SEs, parallel-trends event study, heterogeneity) — players learn that the right number is rarely the first number, that statistical precision is not causal credibility, and that the credibility-revolution framing 'best estimate under stated assumption' beats 'true causal effect'.
Bilingual EN/ES.
Who it is for
An advanced simulation for participants used to working with the main frameworks and trade-offs of the subject, designed for executive education and experienced professionals.
How a session runs
- The instructor creates a session from the Eureka dashboard and invites the participants.
- Participants play 4 rounds. In each one they submit their decisions and the simulation calculates the results. The first round opens with: Estimate naively.
- The simulation can be run in English, Español.
- The instructor follows each participant's progress and results from the dashboard, and uses the class results for the debrief.