synthetic-control

Synthetic control estimates causal effects by constructing a data-driven counterfactual from a donor pool. It optimizes donor weights to match the treated unit's pre-treatment trajectories, allowing you to measure policy impact when randomized experiments are infeasible.

Yes, synthetic control supports multi-unit scenarios including staggered adoption and synthetic DID. It constructs counterfactuals for multiple treated units simultaneously, applying placebo-based inference to ensure robust causal effect estimation across different treatment timings.

Placebo tests for synthetic control inference are run by iteratively assigning the treatment status to each donor unit. This generates a distribution of placebo effects, allowing you to benchmark the actual treatment effect against random permutations and calculate pre/post RMSPE.

Validating a synthetic counterfactual requires checking predictor balance, donor weights, and pre/post RMSPE. These diagnostics ensure the donor pool accurately mirrors the treated unit's pre-treatment characteristics and measure the causal effect's magnitude and significance.

Augmented synthetic control or synthetic DID is used when standard SCM struggles with poor pre-treatment fit or staggered adoption. These approaches improve counterfactual construction by incorporating model adjustments, yielding more credible causal effect estimates.