ARE alumna Rocio Titiunik makes methodological advance in synthetic control settings

ARE alumna and Princeton Professor Rocio Titiunik, along with co-authors, address the challenge of conducting inference in synthetic control settings with staggered treatment adoption. They develop Synthetic Control Prediction Intervals (SCPI) to quantify both in- and out-of-sample uncertainty, providing valid confidence intervals for treatment effects. The paper forthcoming in the Review of Economics and Statistics extends the traditional synthetic control framework by introducing alternative weighting constraints, allowing for treatment anticipation, and enabling inference at both the unit-time and aggregated unit levels. This contribution fills a critical methodological gap, as prior approaches offered limited or no valid inference procedures for staggered adoption designs. An accompanying Journal of Statistical Software paper published Stata, R, and Python packages that implement the proposed methods.