Files
Abstract
We calibrate Linear and Mixed Integer Programs with a bi-level estimator, minimizing under First-order-con-ditions (FOC) conditions a penalty function consider-ing the calibration fit and deviations from given param-eters. To deal with non-convexity, a heuristic generates restart points from current best-fit parameters and their means. Monte-Carlo analysis assesses the ap-proach by drawing parameters for a model optimizing acreages under maximal crop shares, a land balance and annual plus intra-annual labour constraints; a variant comprises integer based investments. Resulting optimal solutions perturbed by white noise provide cal-ibration targets. The approach recovers the true pa-rameters and thus allows for systematic and automated calibration.