Primary Submission Category: Causal Inference and SUTVA/Consistencies Violations
Fuzzy Difference-in-Differences for Spatiotemporal Data
Authors: Andrej Srakar, Marilena Vecco,
Presenting Author: Andrej Srakar*
Difference-in-differences (DiD) literature is a fast growing field in econometrics. Topics such as presence of serial correlation, clustered standard errors, arbitrary covariance structures, parallel growth assumption, negative controls, matching, synthetic control, semi- and nonparametrics, nonlinear models, multiple and continuous treatments and staggered treatment adoption have been subject to recent research. In previous contributions DiD has been extended to spatial data (Delgado and Florax, 2015). We extend this in a treatment effect with network interference context by controlling for violated stable unit treatment values assumption (SUTVA), inherent for such analysis but seldom controlled so far and in a spatiotemporal autoregressive setting. We develop a time-corrected Wald ratio DiD estimator combining fuzzy DiD approach with extensions to changes-in-changes estimation (Athey and Imbens, 2006; De Chaisemartin and D‘Haultfoueille, 2017) with a graphon random graph specification (Li and Wager, 2022). This allows asymptotic analysis in terms of bounds on the moments and Stein approaches including Monte Carlo simulation results. In an application, we study causal effects of the yearly Venice carnival, being able to isolate the effect respective to other competing large events in Venice in the studied period. We consider extensions using spillover double robust DiD and Bayesian approaches.