Distributionally-Robust Price-Elasticity Inference for Computer Agriculture Input Vouchers Under Redemption Fraud: An Artificial Intelligence Minimax Regret Evaluation

Price-elasticity estimates for agricultural input vouchers are routinely used to set subsidy levels, yet redemption fraud contaminates the observed redemption series and biases the elasticity inference that those levels depend on. The elasticity problem is formalised as a partially identified moment condition in which the contamination radius is the adversary's budget; a sharp identification region is derived, and a regret criterion is defined over the ambiguity set. Because no observational voucher data are available for the present analysis, the operating characteristics of the procedure are characterised by a replication-based evaluation of a before-after-control-impact design under a declared variance structure (between-cluster SD = 6.0, residual SD = 4.5, intraclass correlation = 0.64). Power was estimated across 2000 design cells spanning cluster counts of 8, 16, 24 and 32 and effects of 3, 5 and 8 percentage points. The design attains 80% power at 54 clusters for a 5-percentage-point effect.