Publication Date: April 1987
This paper develops a semiparametric method for estimating the nonrandom part V( ) of a random utility function U(v,ω) – V(v) + e(ω) from data on discrete choice behavior. Here v and ω are, respectively, vectors of observable and unobservable attributes of an alternative, and e(ω) is the random part of the utility for that alternative. The method is semiparametric because it assumes that the distribution of the random parts is know up to a ﬁnite-dimensional parameter θ, while not requiring speciﬁcation of a parametric form for V( ).
The nonstochastic part V( ) of the utility function U( ) is assumed to be Lipschitzian and to possess a set of properties, typically assumed for utility functions. The estimator of the pair (V,θ) is shown to be strongly consistent.
Discrete choice models, Nonparametric estimation, Utility functions, Consistency, Semiparametric estimation
JEL Classification Codes: 211, 212, 022
See CFP: 795