Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
Abstract This paper develops mixed integer linear programming (MILP) formulations to compute various revealed preference goodness-of-fit measures. We provide MILP formulations to compute the Houtman–Maks index, the average Varian index, and the minimum cost index when there are linear budgets. Next, we provide MILPs to compute minimal “measurement error” in expenditures, prices, and quantities. Finally, we extend our results to non-linear budgets. As a proof of concept, we compute various goodness-of-fit measures for experimental choice data sets from the literature. The maximal computation time is less than 3 s for all measures examined on these datasets.