Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
This paper introduces new results on the nonparametric identification of separable and nonseparable discrete choice models. It presents constructive methods for recovering the derivatives of the utility functions of the alternatives in a set, when these utility functions are nonparametric and nonseparable in unobservable random terms. When the utility functions are separable, the constructive methods require fewer assumptions. It is assumed that only the probability of choosing one alternative outside the set is observed. The conditions for identification involve testable shape restrictions on the distributions of the nonseparable unobservable random terms.