CAUSAL INFERENCE WORKSHOP 2023: The Mixtape by Scott Cunningham

21 and 22 of September, 2023

UB School of Economics – Facultat d’Economia i Empresa Universitat de Barcelona

The next 21 and 22 of September, the GiM research group and the Observatory on Analysis and Evaluation of Public Policies (OAP) will be hosting Professor Scott Cunningham, professor of economics at Baylor University (Texas) and author of «Causal Inference: The mixtape» who will be giving a 2-days workshop on causal inference. In particular, we will work on advanced methods in difference in differences and synthetic control methods.

Organitzadors: OAP & GiM

Observatori d’Anàlisi i Avaluació de Polítiques Públiques (OAP) & 
Grup de Recerca en Governs i Mercats (GiM)

Registration required

See here (the number of spaces is limited and the Organization will prioritize registration from profiles who might benefit more from this workshop, particularly, PhD candidates and post-docs). The Workshop will be only in person.

panish Ministry of Science and Innovation (PID2022-138866OB-I00)

Financial support

Departament de Recerca Generalitat de Catalunya
Observatori d’Anàlisi i Avaluació de Polítiques Públiques (OAP)

Program

The workshop, that will be only in person, will include practicing exercises with STATA. Bring your laptop.

Check the main readings

Session Thursday 21/9 – Advanced Difference-in-difference (Room: To Be Confirmed) 

9:00 – 18:00

  • Review: Potential outcomes, ATT, DiD and parallel trends.
  • DiD & Covariates: Covariates as correcting for parallel trends violations (Abadie 2005; Heckman, Ichimura and Todd 1997; Sant’Anna and Zhao 2020)
  • DiD & Differential timing: Reviewing the pathologies of two-way fixed effects with differential timing and correcting for it with new estimators (Goodman-Bacon, 2021, Callaway and Sant’Anna 2021; Sun and Abraham 2021; Borusyak, et al. 2023; Chaisemartin and D’Haultfoielle 2020).

Session Friday 22/9 – Synthetic Control (Room: To Be Confirmed) 

9:00 – 13:00

  • Non-negative weighting and interpolating with synthetic control (Abadie, et al.)
  • Least negative weighting and extrapolating with synthetic control (Ben-Michael, et al.)
  • Synthetic DiD (Arkhangelsky et al. 2021)