Nonparametric Identification of Discrete Choice Models

In this paper we give simple proofs of identification results in discrete choice models for the case where neither the deterministic part nor the distribution function of the random parts of the utility function is specified parametrically. The regularity conditions imposed are standard, but differ from conditions applied by other researchers, such as Matzkin (1992, 1993).Nonparametric identification; Discrete choice; Random utility models