Donor Quality

Donor quality has two distinct meanings in GeoSC.

The donor stage measures pairwise pre-period similarity. Correlation captures co-movement; RMSE and MAPE capture level error; normalised DTW captures temporal shape. Composite scores are relative within each treated geography and support screening. They do not estimate final counterfactual weights.

Inference measures joint synthetic-control fit. Several individually modest donors can combine well, while a highly correlated donor can receive little weight. Review fitted pre-period error and the SparseSC weight distribution.

Strong screening scores do not establish donor exchangeability. Eligibility also requires no treatment exposure, comparable outcome measurement, and no post-launch shock that differentially moves the donor. Conversely, poor overlap, incomplete metrics, concentrated recommendation weights, or weak joint pre-fit are direct reasons to reconsider the design.

Donor selection must be outcome-blind with respect to the measured post period. Changing the pool after seeing lift invalidates ordinary interpretation unless the search and multiplicity are explicitly modelled.

Adaptive screening thresholds

The current implementation calculates adaptive thresholds from the entire donor source file. Supply only pre-treatment observations to that stage, as described in Evaluate Donors, to prevent post-period data from influencing the screen. The pre-period filter on pairwise metrics alone does not protect threshold initialisation.

Default metric weights are correlation 0.4, RMSE 0.3, MAPE 0.2 and DTW 0.1. When the source outcome coefficient of variation exceeds 1.0, the correlation and RMSE weights become 0.5 and 0.2. These heuristics rank candidates; they are not calibrated probabilities of design validity.